Scholars@Duke
Lawrence Carin

Lawrence Carin

Professor Emeritus of Electrical and Computer Engineering
Pierre R. Lamond Department of Electrical and Computer Engineering
lcarin@ee.duke.edu
Box 90291, Durham, NC 27708-0291
321 Gross Hall, Durham, NC 27708

Scholarly Works - Journal articles

Detecting implicit biases of large language models with Bayesian hypothesis testing.

Journal article Scientific reports · April 2025 Despite the remarkable performance of large language models (LLMs), such as generative pre-trained Transformers (GPTs), across various tasks, they often perpetuate social biases and stereotypes embedded in their training data. In this paper, we introduce a ... Full text Cite

Proactive Pseudo-Intervention: Pre-informed Contrastive Learning For Interpretable Vision Models

Journal article Proceedings of Machine Learning Research · January 1, 2025 Deep neural networks excel at comprehending complex visual signals, delivering on par or even superior performance to that of human experts. However, ad-hoc visual explanations of model decisions often reveal an alarming level of reliance on exploiting non ... Cite

Text Feature Adversarial Learning for Text Generation With Knowledge Transfer From GPT2.

Journal article IEEE Trans Neural Netw Learn Syst · May 2024 Text generation is a key component of many natural language tasks. Motivated by the success of generative adversarial networks (GANs) for image generation, many text-specific GANs have been proposed. However, due to the discrete nature of text, these text ... Full text Link to item Cite

A machine learning algorithm improves the diagnostic accuracy of the histologic component of antibody mediated rejection (AMR-H) in cardiac transplant endomyocardial biopsies.

Journal article Cardiovasc Pathol · 2024 BACKGROUND: Pathologic antibody mediated rejection (pAMR) remains a major driver of graft failure in cardiac transplant patients. The endomyocardial biopsy remains the primary diagnostic tool but presents with challenges, particularly in distinguishing the ... Full text Link to item Cite

A Deep-Learning Algorithm to Predict Short-Term Progression to Geographic Atrophy on Spectral-Domain Optical Coherence Tomography.

Journal article JAMA Ophthalmol · November 1, 2023 IMPORTANCE: The identification of patients at risk of progressing from intermediate age-related macular degeneration (iAMD) to geographic atrophy (GA) is essential for clinical trials aimed at preventing disease progression. DeepGAze is a fully automated a ... Full text Link to item Cite

Deep-Learning-Based Screening and Ancillary Testing for Thyroid Cytopathology.

Journal article Am J Pathol · September 2023 Thyroid cancer is the most common malignant endocrine tumor. The key test to assess preoperative risk of malignancy is cytologic evaluation of fine-needle aspiration biopsies (FNABs). The evaluation findings can often be indeterminate, leading to unnecessa ... Full text Link to item Cite

Learning Hierarchical Document Graphs From Multilevel Sentence Relations.

Journal article IEEE Trans Neural Netw Learn Syst · August 2023 Organizing the implicit topology of a document as a graph, and further performing feature extraction via the graph convolutional network (GCN), has proven effective in document analysis. However, existing document graphs are often restricted to expressing ... Full text Link to item Cite

Differentiable Hierarchical Optimal Transport for Robust Multi-View Learning.

Journal article IEEE transactions on pattern analysis and machine intelligence · June 2023 Traditional multi-view learning methods often rely on two assumptions: ( i) the samples in different views are well-aligned, and ( ii) their representations obey the same distribution in a latent space. Unfortunately, these two assumptions may be questiona ... Full text Cite

Thyroid Cytopathology Cancer Diagnosis from Smartphone Images Using Machine Learning.

Journal article Mod Pathol · June 2023 We examined the performance of deep learning models on the classification of thyroid fine-needle aspiration biopsies using microscope images captured in 2 ways: with a high-resolution scanner and with a mobile phone camera. Our training set consisted of im ... Full text Link to item Cite

Calibration and Uncertainty in Neural Time-to-Event Modeling.

Journal article IEEE Trans Neural Netw Learn Syst · April 2023 Models for predicting the time of a future event are crucial for risk assessment, across a diverse range of applications. Existing time-to-event (survival) models have focused primarily on preserving pairwise ordering of estimated event times (i.e., relati ... Full text Link to item Cite

Vision transformer-based weakly supervised histopathological image analysis of primary brain tumors

Journal article Iscience · January 20, 2023 Diagnosis of primary brain tumors relies heavily on histopathology. Although various computational pathology methods have been developed for automated diagnosis of primary brain tumors, they usually require neuropathologists’ annotation of region of intere ... Full text Cite

Representing Graphs via Gromov-Wasserstein Factorization.

Journal article IEEE transactions on pattern analysis and machine intelligence · January 2023 Graph representation is a challenging and significant problem for many real-world applications. In this work, we propose a novel paradigm called "Gromov-Wasserstein Factorization" (GWF) to learn graph representations in a flexible and interpretable way. Gi ... Full text Cite

Deep Learning-Assisted Detection of Glaucoma Progression in Spectral-Domain OCT.

Journal article Ophthalmol Glaucoma · 2023 PURPOSE: To develop and validate a deep learning (DL) model for detection of glaucoma progression using spectral-domain (SD)-OCT measurements of retinal nerve fiber layer (RNFL) thickness. DESIGN: Retrospective cohort study. PARTICIPANTS: A total of 14 034 ... Full text Link to item Cite

elBERto: Self-supervised commonsense learning for question answering

Journal article Knowledge Based Systems · December 22, 2022 Commonsense question answering requires reasoning about everyday situations and causes and effects implicit in context. Typically, existing approaches first retrieve external evidence and then perform commonsense reasoning using these evidence. In this pap ... Full text Cite

Explainable multiple abnormality classification of chest CT volumes.

Journal article Artificial intelligence in medicine · October 2022 Understanding model predictions is critical in healthcare, to facilitate rapid verification of model correctness and to guard against use of models that exploit confounding variables. We introduce the challenging new task of explainable multiple abnormalit ... Full text Cite

Lesion identification and malignancy prediction from clinical dermatological images.

Journal article Sci Rep · September 23, 2022 We consider machine-learning-based lesion identification and malignancy prediction from clinical dermatological images, which can be indistinctly acquired via smartphone or dermoscopy capture. Additionally, we do not assume that images contain single lesio ... Full text Link to item Cite

Use of Machine Learning-Based Software for the Screening of Thyroid Cytopathology Whole Slide Images.

Journal article Arch Pathol Lab Med · July 1, 2022 CONTEXT.—: The use of whole slide images (WSIs) in diagnostic pathology presents special challenges for the cytopathologist. Informative areas on a direct smear from a thyroid fine-needle aspiration biopsy (FNAB) smear may be spread across a large area com ... Full text Link to item Cite

A Hybrid Human-Machine Learning Approach for Screening Prostate Biopsies Can Improve Clinical Efficiency Without Compromising Diagnostic Accuracy.

Journal article Arch Pathol Lab Med · June 1, 2022 CONTEXT.—: Prostate cancer is a common malignancy, and accurate diagnosis typically requires histologic review of multiple prostate core biopsies per patient. As pathology volumes and complexity increase, new tools to improve the efficiency of everyday pra ... Full text Link to item Cite

An interpretable deep learning workflow for discovering subvisual abnormalities in CT scans of COVID-19 inpatients and survivors

Journal article Nature Machine Intelligence · May 1, 2022 Tremendous efforts have been made to improve diagnosis and treatment of COVID-19, but knowledge on long-term complications is limited. In particular, a large portion of survivors has respiratory complications, but currently, experienced radiologists and st ... Full text Cite

Convolutional neural network to identify symptomatic Alzheimer's disease using multimodal retinal imaging.

Journal article Br J Ophthalmol · March 2022 BACKGROUND/AIMS: To develop a convolutional neural network (CNN) to detect symptomatic Alzheimer's disease (AD) using a combination of multimodal retinal images and patient data. METHODS: Colour maps of ganglion cell-inner plexiform layer (GC-IPL) thicknes ... Full text Open Access Link to item Cite

WAFFLe: Weight Anonymized Factorization for Federated Learning

Journal article IEEE Access · January 1, 2022 In domains where data are sensitive or private, there is great value in methods that can learn in a distributed manner without the data ever leaving the local devices. In light of this need, federated learning has emerged as a popular training paradigm. Ho ... Full text Cite

Use of convolutional neural networks in skin lesion analysis using real world image and non-image data.

Journal article Front Med (Lausanne) · 2022 BACKGROUND: Understanding performance of convolutional neural networks (CNNs) for binary (benign vs. malignant) lesion classification based on real world images is important for developing a meaningful clinical decision support (CDS) tool. METHODS: We deve ... Full text Link to item Cite

SAGES consensus recommendations on an annotation framework for surgical video.

Journal article Surg Endosc · September 2021 BACKGROUND: The growing interest in analysis of surgical video through machine learning has led to increased research efforts; however, common methods of annotating video data are lacking. There is a need to establish recommendations on the annotation of s ... Full text Link to item Cite

In Reply to Prashar and to Savage.

Journal article Academic medicine : journal of the Association of American Medical Colleges · September 2021 Full text Cite

RetiNerveNet: using recursive deep learning to estimate pointwise 24-2 visual field data based on retinal structure.

Journal article Sci Rep · June 15, 2021 Glaucoma is the leading cause of irreversible blindness in the world, affecting over 70 million people. The cumbersome Standard Automated Perimetry (SAP) test is most frequently used to detect visual loss due to glaucoma. Due to the SAP test's innate diffi ... Full text Link to item Cite

Weakly supervised instance learning for thyroid malignancy prediction from whole slide cytopathology images.

Journal article Med Image Anal · January 2021 We consider machine-learning-based thyroid-malignancy prediction from cytopathology whole-slide images (WSI). Multiple instance learning (MIL) approaches, typically used for the analysis of WSIs, divide the image (bag) into patches (instances), which are u ... Full text Open Access Link to item Cite

Machine-learning-based multiple abnormality prediction with large-scale chest computed tomography volumes.

Journal article Med Image Anal · January 2021 Machine learning models for radiology benefit from large-scale data sets with high quality labels for abnormalities. We curated and analyzed a chest computed tomography (CT) data set of 36,316 volumes from 19,993 unique patients. This is the largest multip ... Full text Link to item Cite

Counterfactual Representation Learning with Balancing Weights

Journal article Proceedings of Machine Learning Research · January 1, 2021 A key to causal inference with observational data is achieving balance in predictive features associated with each treatment type. Recent literature has explored representation learning to achieve this goal. In this work, we discuss the pitfalls of these s ... Cite

Wasserstein contrastive representation distillation

Journal article Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition · January 1, 2021 The primary goal of knowledge distillation (KD) is to encapsulate the information of a model learned from a teacher network into a student network, with the latter being more compact than the former. Existing work, e.g., using Kullback-Leibler divergence f ... Full text Cite

On Artificial Intelligence and Deep Learning Within Medical Education.

Journal article Academic medicine : journal of the Association of American Medical Colleges · November 2020 The methodology of deep learning, a component of machine learning and artificial intelligence, is introduced. The opportunity for this technology to automate some aspects of medical practice is reviewed. Finally, a discussion is provided on the integration ... Full text Cite

Application of a machine learning algorithm to predict malignancy in thyroid cytopathology.

Journal article Cancer Cytopathol · April 2020 BACKGROUND: The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) comprises 6 categories used for the diagnosis of thyroid fine-needle aspiration biopsy (FNAB). Each category has an associated risk of malignancy, which is important in the manage ... Full text Link to item Cite

Digital technology and COVID-19.

Journal article Nat Med · April 2020 The past decade has allowed the development of a multitude of digital tools. Now they can be used to remediate the COVID-19 outbreak. ... Full text Link to item Cite

Artificial Intelligence Mapping of Structure to Function in Glaucoma.

Journal article Transl Vis Sci Technol · March 2020 PURPOSE: To develop an artificial intelligence (AI)-based structure-function (SF) map relating retinal nerve fiber layer (RNFL) damage on spectral domain optical coherence tomography (SDOCT) to functional loss on standard automated perimetry (SAP). METHODS ... Full text Link to item Cite

Survival cluster analysis

Journal article ACM Chil 2020 Proceedings of the 2020 ACM Conference on Health Inference and Learning · February 4, 2020 Conventional survival analysis approaches estimate risk scores or individualized time-to-event distributions conditioned on covariates. In practice, there is often great population-level phenotypic heterogeneity, resulting from (unknown) subpopulations wit ... Full text Cite

Learning autoencoders with relational regularization

Journal article 37th International Conference on Machine Learning Icml 2020 · January 1, 2020 A new algorithmic framework is proposed for learning autoencoders of data distributions. We minimize the discrepancy between the model and target distributions, with a relational regularization on the learnable latent prior. This regularization penalizes t ... Cite

Students Need More Attention: BERT-based Attention Model for Small Data with Application to Automatic Patient Message Triage

Journal article Proceedings of Machine Learning Research · January 1, 2020 Small and imbalanced datasets commonly seen in healthcare represent a challenge when training classifiers based on deep learning models. So motivated, we propose a novel framework based on BioBERT (Bidirectional Encoder Representations from Transformers fo ... Cite

Weakly supervised cross-domain alignment with optimal transport

Journal article 31st British Machine Vision Conference Bmvc 2020 · January 1, 2020 Cross-domain alignment between image objects and text sequences is key to many visual-language tasks, and it poses a fundamental challenge to both computer vision and natural language processing. This paper investigates a novel approach for the identificat ... Cite

Identifying Smoking Environments From Images of Daily Life With Deep Learning.

Journal article JAMA Netw Open · August 2, 2019 IMPORTANCE: Environments associated with smoking increase a smoker's craving to smoke and may provoke lapses during a quit attempt. Identifying smoking risk environments from images of a smoker's daily life provides a basis for environment-based interventi ... Full text Link to item Cite

A convergence analysis for a class of practical variance-reduction stochastic gradient MCMC

Journal article Science China Information Sciences · January 1, 2019 Stochastic gradient Markov chain Monte Carlo (SG-MCMC) has been developed as a flexible family of scalable Bayesian sampling algorithms. However, there has been little theoretical analysis of the impact of minibatch size to the algorithm’s convergence rate ... Full text Cite

Continuing progress of spike sorting in the era of big data

Journal article Current opinion in neurobiology · 2019 Cite

Improving textual network learning with variational homophilic embeddings

Journal article Advances in Neural Information Processing Systems · January 1, 2019 The performance of many network learning applications crucially hinges on the success of network embedding algorithms, which aim to encode rich network information into low-dimensional vertex-based vector representations. This paper considers a novel varia ... Cite

Thyroid Cancer Malignancy Prediction From Whole Slide Cytopathology Images

Journal article Proceedings of Machine Learning Research · January 1, 2019 We consider preoperative prediction of thyroid cancer based on ultra-high-resolution whole-slide cytopathology images. Inspired by how human experts perform diagnosis, our approach first identifies and classifies diagnostic image regions containing informa ... Cite

Kernel-based approaches for sequence modeling: Connections to neural methods

Journal article Advances in Neural Information Processing Systems · January 1, 2019 We investigate time-dependent data analysis from the perspective of recurrent kernel machines, from which models with hidden units and gated memory cells arise naturally. By considering dynamic gating of the memory cell, a model closely related to the long ... Cite

On Deep Learning for Medical Image Analysis.

Journal article JAMA · September 18, 2018 Full text Link to item Cite

Adversarial Time-to-Event Modeling.

Journal article Proc Mach Learn Res · July 2018 Modern health data science applications leverage abundant molecular and electronic health data, providing opportunities for machine learning to build statistical models to support clinical practice. Time-to-event analysis, also called survival analysis, st ... Link to item Cite

Brain-wide Electrical Spatiotemporal Dynamics Encode Depression Vulnerability.

Journal article Cell · March 22, 2018 Brain-wide fluctuations in local field potential oscillations reflect emergent network-level signals that mediate behavior. Cracking the code whereby these oscillations coordinate in time and space (spatiotemporal dynamics) to represent complex behaviors w ... Full text Link to item Cite

Deconvolutional latent-variable model for text sequence matching

Journal article 32nd Aaai Conference on Artificial Intelligence Aaai 2018 · January 1, 2018 A latent-variable model is introduced for text matching, inferring sentence representations by jointly optimizing generative and discriminative objectives. To alleviate typical optimization challenges in latent-variable models for text, we employ deconvolu ... Cite

Multi-Label Learning from Medical Plain Text with Convolutional Residual Models

Journal article Proceedings of Machine Learning Research · January 1, 2018 Predicting diagnoses from Electronic Health Records (EHRs) is an important medical application of multi-label learning. We propose a convolutional residual model for multi-label classification from doctor notes in EHR data. A given patient may have multipl ... Cite

Adversarial time-to-event modeling

Journal article 35th International Conference on Machine Learning Icml 2018 · January 1, 2018 Modern health data science applications leverage abundant molecular and electronic health data; providing opportunities for machine learning to build statistical models to support clinical practice. Time-to-event analysis, also called survival analysis, st ... Cite

Joint embedding of words and labels for text classification

Journal article Acl 2018 56th Annual Meeting of the Association for Computational Linguistics Proceedings of the Conference Long Papers · January 1, 2018 Word embeddings are effective intermediate representations for capturing semantic regularities between words, when learning the representations of text sequences. We propose to view text classification as a label-word joint embedding problem: each label is ... Full text Cite

NasH: Toward end-to-end neural architecture for generative semantic hashing

Journal article Acl 2018 56th Annual Meeting of the Association for Computational Linguistics Proceedings of the Conference Long Papers · January 1, 2018 Semantic hashing has become a powerful paradigm for fast similarity search in many information retrieval systems. While fairly successful, previous techniques generally require two-stage training, and the binary constraints are handled ad-hoc. In this pape ... Full text Open Access Cite

Baseline needs more love: On simple word-embedding-based models and associated pooling mechanisms

Journal article Acl 2018 56th Annual Meeting of the Association for Computational Linguistics Proceedings of the Conference Long Papers · January 1, 2018 Many deep learning architectures have been proposed to model the compositionality in text sequences, requiring a substantial number of parameters and expensive computations. However, there has not been a rigorous evaluation regarding the added value of sop ... Full text Cite

JointGAN: Multi-domain joint distribution learning with generative adversarial nets

Journal article 35th International Conference on Machine Learning Icml 2018 · January 1, 2018 A new generative adversarial network is developed for joint distribution matching. Distinct from most existing approaches, that only learn conditional distributions, the proposed model aims to learn a joint distribution of multiple random variables (domain ... Cite

Improved semantic-aware network embedding with fine-grained word alignment

Journal article Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing Emnlp 2018 · January 1, 2018 Network embeddings, which learn low-dimensional representations for each vertex in a large-scale network, have received considerable attention in recent years. For a wide range of applications, vertices in a network are typically accompanied by rich textua ... Full text Cite

Dynamically Timed Stimulation of Corticolimbic Circuitry Activates a Stress-Compensatory Pathway.

Journal article Biol Psychiatry · December 15, 2017 BACKGROUND: The prefrontal cortex plays a critical role in regulating emotional behaviors, and dysfunction of prefrontal cortex-dependent networks has been broadly implicated in mediating stress-induced behavioral disorders including major depressive disor ... Full text Link to item Cite

Information-Theoretic Compressive Measurement Design.

Journal article IEEE transactions on pattern analysis and machine intelligence · June 2017 An information-theoretic projection design framework is proposed, of interest for feature design and compressive measurements. Both Gaussian and Poisson measurement models are considered. The gradient of a proposed information-theoretic metric (ITM) is der ... Full text Cite

Adversarial feature matching for text generation

Journal article 34th International Conference on Machine Learning Icml 2017 · January 1, 2017 The Generative Adversarial Network (GAN) has achieved great success in generating realistic (real-valued) synthetic data. However, convergence issues and difficulties dealing with discrete data hinder the applicability of GAN to text. We propose a framewor ... Cite

Adversarial symmetric variational autoencoder

Journal article Advances in Neural Information Processing Systems · January 1, 2017 A new form of variational autoencoder (VAE) is developed, in which the joint distribution of data and codes is considered in two (symmetric) forms: (i) from observed data fed through the encoder to yield codes, and (ii) from latent codes drawn from a simpl ... Cite

Deconvolutional paragraph representation learning

Journal article Advances in Neural Information Processing Systems · January 1, 2017 Learning latent representations from long text sequences is an important first step in many natural language processing applications. Recurrent Neural Networks (RNNs) have become a cornerstone for this challenging task. However, the quality of sentences du ... Cite

Stochastic gradient monomial gamma sampler

Journal article 34th International Conference on Machine Learning Icml 2017 · January 1, 2017 Recent advances in stochastic gradient techniques have made it possible to estimate posterior distributions from large datasets via Markov Chain Monte Carlo (MCMC). However, when the target posterior is multimodal, mixing performance is often poor. This re ... Cite

Computational Snapshot Multispectral Cameras: Toward dynamic capture of the spectral world

Journal article IEEE Signal Processing Magazine · September 1, 2016 Multispectral cameras collect image data with a greater number of spectral channels than traditional trichromatic sensors, thus providing spectral information at a higher level of detail. Such data are useful in various fields, such as remote sensing, mate ... Full text Cite

Efficient patch-based approach for compressive depth imaging.

Journal article Applied optics · September 2016 We present efficient camera hardware and algorithms to capture images with extended depth of field. The camera moves its focal plane via a liquid lens and modulates the scene at different focal planes by shifting a fixed binary mask, with synchronization a ... Full text Cite

Dysregulation of Prefrontal Cortex-Mediated Slow-Evolving Limbic Dynamics Drives Stress-Induced Emotional Pathology.

Journal article Neuron · July 20, 2016 Circuits distributed across cortico-limbic brain regions compose the networks that mediate emotional behavior. The prefrontal cortex (PFC) regulates ultraslow (<1 Hz) dynamics across these networks, and PFC dysfunction is implicated in stress-related illne ... Full text Link to item Cite

Electronic health record analysis via deep poisson factor models

Journal article Journal of Machine Learning Research · April 1, 2016 Electronic Health Record (EHR) phenotyping utilizes patient data captured through normal medical practice, to identify features that may represent computational medical phenotypes. These features may be used to identify at-risk patients and improve predict ... Cite

Host gene expression classifiers diagnose acute respiratory illness etiology.

Journal article Sci Transl Med · January 20, 2016 Acute respiratory infections caused by bacterial or viral pathogens are among the most common reasons for seeking medical care. Despite improvements in pathogen-based diagnostics, most patients receive inappropriate antibiotics. Host response biomarkers of ... Full text Open Access Link to item Cite

Applying compressive sensing to TEM video: a substantial frame rate increase on any camera

Journal article Advanced Structural and Chemical Imaging · December 1, 2015 One of the main limitations of imaging at high spatial and temporal resolution during in-situ transmission electron microscopy (TEM) experiments is the frame rate of the camera being used to image the dynamic process. While the recent development of direct ... Full text Cite

Alternating minimization algorithm with automatic relevance determination for transmission tomography under poisson noise

Journal article SIAM Journal on Imaging Sciences · September 30, 2015 We propose a globally convergent alternating minimization (AM) algorithm for image reconstruction in transmission tomography, which extends automatic relevance determination (ARD) to Poisson noise models with Beer’s law. The algorithm promotes solutions th ... Full text Cite

Image translation for single-shot focal tomography

Journal article Optica · September 20, 2015 Focus and depth of field are conventionally addressed by adjusting longitudinal lens position. More recently, combinations of deliberate blur and computational processing have been used to extend depth of field. Here we show that dynamic control of transve ... Full text Cite

Signal recovery and system calibration from multiple compressive poisson measurements

Journal article SIAM Journal on Imaging Sciences · September 17, 2015 The measurement matrix employed in compressive sensing typically cannot be known precisely a priori and must be estimated via calibration. One may take multiple compressive measurements, from which the measurement matrix and underlying signals may be estim ... Full text Cite

Compressive hyperspectral imaging with side information

Journal article IEEE Journal on Selected Topics in Signal Processing · September 1, 2015 A blind compressive sensing algorithm is proposed to reconstruct hyperspectral images from spectrally-compressed measurements. The wavelength-dependent data are coded and then superposed, mapping the three-dimensional hyperspectral datacube to a two-dimens ... Full text Cite

Spectral-temporal compressive imaging.

Journal article Optics letters · September 2015 This Letter presents a compressive camera that integrates mechanical translation and spectral dispersion to compress a multi-spectral, high-speed scene onto a monochrome, video-rate detector. Experimental reconstructions of 17 spectral channels and 11 temp ... Full text Cite

Multivariate time-series analysis and diffusion maps

Journal article Signal Processing · April 25, 2015 Dimensionality reduction in multivariate time series analysis has broad applications, ranging from financial data analysis to biomedical research. However, high levels of ambient noise and various interferences result in nonstationary signals, which may le ... Full text Cite

Erratum: Finite sample posterior concentration in high-dimensional regression (Information and Inference (2015) 3 (103-133) DOI: 10.1093/imaiai/iau003)

Journal article Information and Inference · March 1, 2015 Artin Armagan's and Rayan Saab's affiliations were switched in the published version of this article. Artin Armagan's affiliation should be: SAS Institute, Inc., Raleigh, NC, USA; Rayan Saab's affiliation should be: Department of Mathematics, University of ... Full text Cite

Negative Binomial Process Count and Mixture Modeling.

Journal article IEEE transactions on pattern analysis and machine intelligence · February 2015 The seemingly disjoint problems of count and mixture modeling are united under the negative binomial (NB) process. A gamma process is employed to model the rate measure of a Poisson process, whose normalization provides a random probability measure for mix ... Full text Cite

A Bayesian Nonparametric Approach to Image Super-Resolution.

Journal article IEEE transactions on pattern analysis and machine intelligence · February 2015 Super-resolution methods form high-resolution images from low-resolution images. In this paper, we develop a new Bayesian nonparametric model for super-resolution. Our method uses a beta-Bernoulli process to learn a set of recurring visual patterns, called ... Full text Cite

Quantitative arbor analytics: unsupervised harmonic co-clustering of populations of brain cell arbors based on L-measure.

Journal article Neuroinformatics · January 2015 This paper presents a robust unsupervised harmonic co-clustering method for profiling arbor morphologies for ensembles of reconstructed brain cells (e.g., neurons, microglia) based on quantitative measurements of the cellular arbors. Specifically, this met ... Full text Cite

Compressive sensing by learning a Gaussian mixture model from measurements.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · January 2015 Compressive sensing of signals drawn from a Gaussian mixture model (GMM) admits closed-form minimum mean squared error reconstruction from incomplete linear measurements. An accurate GMM signal model is usually not available a priori, because it is difficu ... Full text Cite

Non-Gaussian discriminative factor models via the max-margin rank-likelihood

Journal article 32nd International Conference on Machine Learning Icml 2015 · January 1, 2015 We consider the problem of discriminative factor analysis for data that are in general non-Gaussian. A Bayesian model based on the ranks of the data is proposed. We first introduce a new max-margin version of the rank-likelihood. A discriminative factor mo ... Open Access Cite

Deep temporal sigmoid belief networks for sequence modeling

Journal article Advances in Neural Information Processing Systems · 2015 Cite

Estimation of the CSA-ODF using Bayesian compressed sensing of multi-shell HARDI.

Journal article Magnetic resonance in medicine · November 2014 PurposeDiffusion MRI provides important information about the brain white matter structures and has opened new avenues for neuroscience and translational research. However, acquisition time needed for advanced applications can still be a challenge ... Full text Cite

Video compressive sensing using Gaussian mixture models.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · November 2014 A Gaussian mixture model (GMM)-based algorithm is proposed for video reconstruction from temporally compressed video measurements. The GMM is used to model spatio-temporal video patches, and the reconstruction can be efficiently computed based on analytic ... Full text Cite

Hierarchical Infinite Divisibility for Multiscale Shrinkage

Journal article IEEE Transactions on Signal Processing · September 1, 2014 A new shrinkage-based construction is developed for a compressible vector x e ℝn, for cases in which the components of are naturally associated with a tree structure. Important examples are when corresponds to the coefficients of a wavelet or bl ... Full text Cite

Compressed sampling strategies for tomography.

Journal article Journal of the Optical Society of America. A, Optics, image science, and vision · July 2014 We investigate new sampling strategies for projection tomography, enabling one to employ fewer measurements than expected from classical sampling theory without significant loss of information. Inspired by compressed sensing, our approach is based on the u ... Full text Cite

Off-policy reinforcement learning with Gaussian processes

Journal article IEEE Caa Journal of Automatica Sinica · July 1, 2014 An off-policy Bayesian nonparameteric approximate reinforcement learning framework, termed as GPQ, that employs a Gaussian processes (GP) model of the value (Q) function is presented in both the batch and online settings. Sufficient conditions on GP hyperp ... Full text Cite

Generalized alternating projection for weighted-l2,1 minimization with applications to model-based compressive sensing

Journal article SIAM Journal on Imaging Sciences · June 26, 2014 We consider the group basis pursuit problem, which extends basis pursuit by replacing the l1 norm with a weighted-L2,1 norm. We provide an anytime algorithm, called generalized alternating projection (GAP), to solve this problem. The ... Full text Cite

Finite sample posterior concentration in high-dimensional regression

Journal article Information and Inference · June 1, 2014 We study the behavior of the posterior distribution in high-dimensional Bayesian Gaussian linear regression models having p ≫ n, where p is the number of predictors and n is the sample size. Our focus is on obtaining quantitative finite sample bounds ensur ... Full text Cite

Bayesian joint analysis of heterogeneous genomics data.

Journal article Bioinformatics (Oxford, England) · May 2014 SummaryA non-parametric Bayesian factor model is proposed for joint analysis of multi-platform genomics data. The approach is based on factorizing the latent space (feature space) into a shared component and a data-specific component with the dime ... Full text Cite

Reconstruction of signals drawn from a gaussian mixture via noisy compressive measurements

Journal article IEEE Transactions on Signal Processing · May 1, 2014 This paper determines to within a single measurement the minimum number of measurements required to successfully reconstruct a signal drawn from a Gaussian mixture model in the low-noise regime. The method is to develop upper and lower bounds that are a fu ... Full text Cite

The potential for Bayesian compressive sensing to significantly reduce electron dose in high-resolution STEM images.

Journal article Microscopy (Oxford, England) · February 2014 The use of high-resolution imaging methods in scanning transmission electron microscopy (STEM) is limited in many cases by the sensitivity of the sample to the beam and the onset of electron beam damage (for example, in the study of organic systems, in tom ... Full text Cite

Compressive coded aperture spectral imaging: An introduction

Journal article IEEE Signal Processing Magazine · January 1, 2014 Maging spectroscopy involves the sensing of a large amount of spatial information across a multitude of wavelengths. Conventional approaches to hyperspectral sensing scan adjacent zones of the underlying spectral scene and merge the results to construct a ... Full text Cite

An active learning approach for rapid characterization of endothelial cells in human tumors.

Journal article PloS one · January 2014 Currently, no available pathological or molecular measures of tumor angiogenesis predict response to antiangiogenic therapies used in clinical practice. Recognizing that tumor endothelial cells (EC) and EC activation and survival signaling are the direct t ... Full text Open Access Cite

Bayesian modeling of temporal properties of infectious disease in a college student population

Journal article Journal of Applied Statistics · January 1, 2014 A Bayesian statistical model is developed for analysis of the time-evolving properties of infectious disease, with a particular focus on viruses. The model employs a latent semi-Markovian state process, and the state-transition statistics are driven by thr ... Full text Cite

An integrated transcriptome and expressed variant analysis of sepsis survival and death.

Journal article Genome Med · 2014 BACKGROUND: Sepsis, a leading cause of morbidity and mortality, is not a homogeneous disease but rather a syndrome encompassing many heterogeneous pathophysiologies. Patient factors including genetics predispose to poor outcomes, though current clinical ch ... Full text Open Access Link to item Cite

Foreword

Journal article · 2014 Open Access Cite

Online expectation maximization for reinforcement learning in POMDPs

Journal article Ijcai International Joint Conference on Artificial Intelligence · December 1, 2013 We present online nested expectation maximization for model-free reinforcement learning in a POMDP. The algorithm evaluates the policy only in the current learning episode, discarding the episode after the evaluation and memorizing the sufficient statistic ... Cite

Reconstruction of Gaussian mixture models from compressive measurements: A phase transition view

Journal article 2013 IEEE Global Conference on Signal and Information Processing Globalsip 2013 Proceedings · December 1, 2013 We characterize the minimum number of measurements needed to drive to zero the minimum mean squared error (MMSE) of Gaussian mixture model (GMM) input signals in the low-noise regime. The result also hints at almost phase-transition optimal recovery proced ... Full text Cite

Compressive sensing for incoherent imaging systems with optical constraints

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · October 18, 2013 We consider the problem of linear projection design for incoherent optical imaging systems. We propose a computationally efficient method to obtain effective measurement kernels that satisfy the physical constraints imposed by an optical system, starting f ... Full text Cite

A host-based RT-PCR gene expression signature to identify acute respiratory viral infection.

Journal article Sci Transl Med · September 18, 2013 Improved ways to diagnose acute respiratory viral infections could decrease inappropriate antibacterial use and serve as a vital triage mechanism in the event of a potential viral pandemic. Measurement of the host response to infection is an alternative to ... Full text Link to item Cite

Analysis of space-time relational data with application to legislative voting

Journal article Computational Statistics and Data Analysis · July 29, 2013 We consider modeling spatio-temporally indexed relational data, motivated by analysis of voting data for the United States House of Representatives over two decades. The data are characterized by incomplete binary matrices, representing votes of legislator ... Full text Cite

An integrated clinico-metabolomic model improves prediction of death in sepsis.

Journal article Sci Transl Med · July 24, 2013 Sepsis is a common cause of death, but outcomes in individual patients are difficult to predict. Elucidating the molecular processes that differ between sepsis patients who survive and those who die may permit more appropriate treatments to be deployed. We ... Full text Link to item Cite

Coded hyperspectral imaging and blind compressive sensing

Journal article SIAM Journal on Imaging Sciences · July 15, 2013 Blind compressive sensing (CS) is considered for reconstruction of hyperspectral data imaged by a coded aperture camera. The measurements are manifested as a superposition of the coded wavelength-dependent data, with the ambient three-dimensional hyperspec ... Full text Cite

Bayesian Gaussian Copula Factor Models for Mixed Data.

Journal article Journal of the American Statistical Association · June 2013 Gaussian factor models have proven widely useful for parsimoniously characterizing dependence in multivariate data. There is a rich literature on their extension to mixed categorical and continuous variables, using latent Gaussian variables or through gene ... Full text Open Access Cite

Latent protein trees

Journal article Annals of Applied Statistics · June 1, 2013 Unbiased, label-free proteomics is becoming a powerful technique for measuring protein expression in almost any biological sample. The output of these measurements after preprocessing is a collection of features and their associated intensities for each sa ... Full text Open Access Cite

Coded aperture compressive temporal imaging.

Journal article Optics express · May 2013 We use mechanical translation of a coded aperture for code division multiple access compression of video. We discuss the compressed video's temporal resolution and present experimental results for reconstructions of > 10 frames of temporal data per coded s ... Full text Open Access Cite

Spatio-temporal modeling of legislation and votes

Journal article Bayesian Analysis · March 22, 2013 A model is presented for analysis of multivariate binary data with spatio-temporal dependencies, and applied to congressional roll call data from the United States House of Representatives and Senate. The model considers each legislator's constituency (loc ... Full text Cite

Task-driven adaptive statistical compressive sensing of gaussian mixture models

Journal article IEEE Transactions on Signal Processing · January 21, 2013 A framework for adaptive and non-adaptive statistical compressive sensing is developed, where a statistical model replaces the standard sparsity model of classical compressive sensing. We propose within this framework optimal task-specific sensing protocol ... Full text Open Access Cite

Patient clustering with uncoded text in electronic medical records.

Journal article AMIA Annu Symp Proc · 2013 We propose a mixture model for text data designed to capture underlying structure in the history of present illness section of electronic medical records data. Additionally, we propose a method to induce bias that leads to more homogeneous sets of diagnose ... Link to item Cite

Online expectation maximization for reinforcement learning in POMDPs

Journal article IJCAI International Joint Conference on Artificial Intelligence · 2013 Cite

Gaussian mixture model for video compressive sensing

Journal article 2013 IEEE International Conference on Image Processing Icip 2013 Proceedings · January 1, 2013 A Gaussian Mixture Model (GMM)-based algorithm is proposed for video reconstruction from temporal compressed measurements. The GMM is used to model spatio-temporal video patches, and the reconstruction can be efficiently computed based on analytic expressi ... Full text Cite

Exploring the mind: Integrating questionnaires and fMRI

Journal article 30th International Conference on Machine Learning Icml 2013 · January 1, 2013 A new model is developed for joint analysis of ordered, categorical, real and count data. The ordered and categorical data are answers to questionnaires, the (word) count data correspond to the text questions from the questionnaires, and the real data corr ... Cite

Compressive sensing for video using a passive coding element

Journal article Optics InfoBase Conference Papers · January 1, 2013 We present a prototype system that utilizes mechanical translation of a passive coding element to compress high-speed temporal information into low-framerate video sequences. Reconstructions of 148 frames per experimental coded snapshot are reported. © OSA ... Cite

Designed measurements for vector count data

Journal article Advances in neural information processing systems · 2013 Cite

Integrated non-factorized variational inference

Journal article Advances in Neural Information Processing Systems · January 1, 2013 We present a non-factorized variational method for full posterior inference in Bayesian hierarchical models, with the goal of capturing the posterior variable dependencies via efficient and possibly parallel computation. Our approach unifies the integrated ... Cite

Real-time inference for a gamma process model of neural spiking

Journal article Advances in Neural Information Processing Systems · 2013 Cite

Dynamic clustering via asymptotics of the dependent Dirichlet process mixture

Journal article Advances in Neural Information Processing Systems · January 1, 2013 This paper presents a novel algorithm, based upon the dependent Dirichlet process mixture model (DDPMM), for clustering batch-sequential data containing an unknown number of evolving clusters. The algorithm is derived via a lowvariance asymptotic analysis ... Cite

Deep Learning with Hierarchical Convolutional Factor Analysis.

Journal article IEEE transactions on pattern analysis and machine intelligence · January 2013 Unsupervised multi-layered ("deep") models are considered for general data, with a particular focus on imagery. The model is represented using a hierarchical convolutional factor-analysis construction, with sparse factor loadings and scores. The computatio ... Cite

A host transcriptional signature for presymptomatic detection of infection in humans exposed to influenza H1N1 or H3N2.

Journal article PLoS One · 2013 There is great potential for host-based gene expression analysis to impact the early diagnosis of infectious diseases. In particular, the influenza pandemic of 2009 highlighted the challenges and limitations of traditional pathogen-based testing for suspec ... Full text Open Access Link to item Cite

Generalized Bregman divergence and gradient of mutual information for vector Poisson channels

Journal article IEEE International Symposium on Information Theory Proceedings · January 1, 2013 We investigate connections between information-theoretic and estimation-theoretic quantities in vector Poisson channel models. In particular, we generalize the gradient of mutual information with respect to key system parameters from the scalar to the vect ... Full text Open Access Cite

Adaptive temporal compressive sensing for video

Journal article 2013 IEEE International Conference on Image Processing Icip 2013 Proceedings · January 1, 2013 This paper introduces the concept of adaptive temporal compressive sensing (CS) for video. We propose a CS algorithm to adapt the compression ratio based on the scene's temporal complexity, computed from the compressed data, without compromising the qualit ... Full text Open Access Cite

Quantitative profiling of microglia populations using harmonic co-clustering of arbor morphology measurements

Journal article Proceedings International Symposium on Biomedical Imaging · January 1, 2013 Microglia are the resident immune cell population in the mammalian central nervous system (CNS). These highly plastic cells exhibit ramified arbors in their resting state, and progressively less-complex arbors when activated. Our goal is to compare the spa ... Full text Cite

Dictionary learning for hyperspectral video compressive sensing

Journal article Frontiers in Optics, FIO 2012 · December 1, 2012 Blind compressive sensing (CS) is considered for reconstruction of hyperspectral data imaged by a coded aperture camera. The measurements are manifested as a superposition of the coded wavelengthdependent data, with the ambient three-dimensional hyperspect ... Cite

Augment-and-conquer negative binomial processes

Journal article Advances in Neural Information Processing Systems · December 1, 2012 By developing data augmentation methods unique to the negative binomial (NB) distribution, we unite seemingly disjoint count and mixture models under the NB process framework. We develop fundamental properties of the models and derive efficient Gibbs sampl ... Open Access Cite

Nested dictionary learning for hierarchical organization of imagery and text

Journal article Uncertainty in Artificial Intelligence Proceedings of the 28th Conference Uai 2012 · December 1, 2012 A tree-based dictionary learning model is developed for joint analysis of imagery and associated text. The dictionary learning may be applied directly to the imagery from patches, or to general feature vectors extracted from patches or superpixels (using a ... Open Access Cite

Nonparametric Bayesian Segmentation of a Multivariate Inhomogeneous Space-Time Poisson Process.

Journal article Bayesian analysis · December 2012 A nonparametric Bayesian model is proposed for segmenting time-evolving multivariate spatial point process data. An inhomogeneous Poisson process is assumed, with a logistic stick-breaking process (LSBP) used to encourage piecewise-constant spatial Poisson ... Full text Cite

Joint modeling of a matrix with associated text via latent binary features

Journal article Advances in Neural Information Processing Systems · December 1, 2012 A new methodology is developed for joint analysis of a matrix and accompanying documents, with the documents associated with the matrix rows/columns. The documents are modeled with a focused topic model, inferring interpretable latent binary features for e ... Cite

Active learning for large-scale factor analysis

Journal article 2012 IEEE Statistical Signal Processing Workshop Ssp 2012 · November 6, 2012 A method for Bayesian factor analysis (FA) of large matrices is proposed. It is assumed that a small number of matrix elements are initially observed, and the statistical FA model is employed to actively and sequentially select which new matrix entries wou ... Full text Cite

How to focus the discriminative power of a dictionary

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · October 23, 2012 This paper is motivated by the challenge of high fidelity processing of images using a relatively small set of projection measurements. This is a problem of great interest in many sensing applications, for example where high photodetector counts are preclu ... Full text Cite

Adapted statistical compressive sensing: Learning to sense gaussian mixture models

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · October 23, 2012 A framework for learning sensing kernels adapted to signals that follow a Gaussian mixture model (GMM) is introduced in this paper. This follows the paradigm of statistical compressive sensing (SCS), where a statistical model, a GMM in particular, replaces ... Full text Cite

Online Bayesian dictionary learning for large datasets

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · October 23, 2012 The problem of learning a data-adaptive dictionary for a very large collection of signals is addressed. This paper proposes a statistical model and associated variational Bayesian (VB) inference for simultaneously learning the dictionary and performing spa ... Full text Cite

Communications inspired linear discriminant analysis

Journal article Proceedings of the 29th International Conference on Machine Learning Icml 2012 · October 10, 2012 We study the problem of supervised linear dimensionality reduction, taking an information-theoretic viewpoint. The linear projection matrix is designed by maximizing the mutual information between the projected signal and the class label. By harnessing a r ... Open Access Cite

Lognormal and gamma mixed negative binomial regression

Journal article Proceedings of the 29th International Conference on Machine Learning Icml 2012 · October 10, 2012 In regression analysis of counts, a lack of simple and efficient algorithms for posterior computation has made Bayesian approaches appear unattractive and thus underdeveloped. We propose a lognormal and gamma mixed negative binomial (NB) regression model f ... Open Access Cite

Cross-domain multitask learning with latent probit models

Journal article Proceedings of the 29th International Conference on Machine Learning Icml 2012 · October 10, 2012 Learning multiple tasks across heterogeneous domains is a challenging problem since the feature space may not be the same for different tasks. We assume the data in multiple tasks are generated from a latent common domain via sparse domain transforms and p ... Open Access Cite

Lévy measure decompositions for the beta and gamma processes

Journal article Proceedings of the 29th International Conference on Machine Learning Icml 2012 · October 10, 2012 We develop new representations for the Lévy measures of the beta and gamma processes. These representations are manifested in terms of an infinite sum of well-behaved (proper) beta and gamma distributions. Further, we demonstrate how these infinite sums ma ... Cite

Inferring latent structure from mixed real and categorical relational data

Journal article Proceedings of the 29th International Conference on Machine Learning Icml 2012 · October 10, 2012 We consider analysis of relational data (a matrix), in which the rows correspond to subjects (e.g., people) and the columns correspond to attributes. The elements of the matrix may be a mix of real and categorical. Each subject and attribute is characteriz ... Open Access Cite

The contextual focused topic model

Journal article Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · September 14, 2012 A nonparametric Bayesian contextual focused topic model (cFTM) is proposed. The cFTM infers a sparse ("focused") set of topics for each document, while also leveraging contextual information about the author(s) and document venue. The hierarchical beta pro ... Full text Cite

Active learning for online bayesian matrix factorization

Journal article Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · September 14, 2012 The problem of large-scale online matrix completion is addressed via a Bayesian approach. The proposed method learns a factor analysis (FA) model for large matrices, based on a small number of observed matrix elements, and leverages the statistical model t ... Full text Cite

Hierarchical factor modeling of proteomics data

Journal article 2012 IEEE 2nd International Conference on Computational Advances in Bio and Medical Sciences Iccabs 2012 · May 8, 2012 This paper presents a hierarchical bayesian factor model specifically designed to model the known correlation structure of both peptides and proteins in unbiased, label free proteomics. The model utilizes partial identification information from peptide seq ... Full text Cite

Dictionary learning for noisy and incomplete hyperspectral images

Journal article SIAM Journal on Imaging Sciences · February 13, 2012 We consider analysis of noisy and incomplete hyperspectral imagery, with the objective of removing the noise and inferring the missing data. The noise statistics may be wavelength dependent, and the fraction of data missing (at random) may be substantial, ... Full text Cite

Nonparametric Bayesian dictionary learning for analysis of noisy and incomplete images.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · January 2012 Nonparametric Bayesian methods are considered for recovery of imagery based upon compressive, incomplete, and/or noisy measurements. A truncated beta-Bernoulli process is employed to infer an appropriate dictionary for the data under test and also for imag ... Full text Cite

Communications-inspired projection design with application to compressive sensing

Journal article SIAM Journal on Imaging Sciences · January 1, 2012 We consider the recovery of an underlying signal x ∈ ℂm based on projection measurements of the form y = Mx+w, where y ∈ ℂℓ and w is measurement noise; we are interested in the case ℓ ≪ m. It is assumed that the signal model p(x) is k ... Full text Open Access Cite

Beta-negative binomial process and poisson factor analysis

Journal article Journal of Machine Learning Research · January 1, 2012 A beta-negative binomial (BNB) process is proposed, leading to a beta-gamma-Poisson process, which may be viewed as a "multiscoop" generalization of the beta-Bernoulli process. The BNB process is augmented into a beta-gamma-gamma-Poisson hierarchical struc ... Open Access Cite

Bayesian robust principal component analysis.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · December 2011 A hierarchical Bayesian model is considered for decomposing a matrix into low-rank and sparse components, assuming the observed matrix is a superposition of the two. The matrix is assumed noisy, with unknown and possibly non-stationary noise statistics. Th ... Full text Cite

The kernel beta process

Journal article Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011 · December 1, 2011 A new Lévy process prior is proposed for an uncountable collection of covariate-dependent feature-learning measures; the model is called the kernel beta process (KBP). Available covariates are handled efficiently via the kernel construction, with covariate ... Cite

Hierarchical topic modeling for analysis of time-evolving personal choices

Journal article Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011 · December 1, 2011 The nested Chinese restaurant process is extended to design a nonparametric topic-model tree for representation of human choices. Each tree path corresponds to a type of person, and each node (topic) has a corresponding probability vector over items that m ... Cite

Learning discriminative sparse representations for modeling, source separation, and mapping of hyperspectral imagery

Journal article IEEE Transactions on Geoscience and Remote Sensing · November 1, 2011 A method is presented for subpixel modeling, mapping, and classification in hyperspectral imagery using learned block-structured discriminative dictionaries, where each block is adapted and optimized to represent a material in a compact and sparse manner. ... Full text Cite

The infinite regionalized policy representation

Journal article Proceedings of the 28th International Conference on Machine Learning Icml 2011 · October 7, 2011 We introduce the infinite regionalized policy presentation (iRPR), as a nonparametric policy for reinforcement learning in partially observable Markov decision processes (POMDPs). The iRPR assumes an unbounded set of decision states a priori, and infers th ... Cite

On the integration of topic modeling and dictionary learning

Journal article Proceedings of the 28th International Conference on Machine Learning Icml 2011 · October 7, 2011 A new nonparametric Bayesian model is developed to integrate dictionary learning and topic model into a unified framework. The model is employed to analyze partially annotated images, with the dictionary learning performed directly on image patches. Effici ... Cite

The hierarchical beta process for convolutional factor analysis and deep learning

Journal article Proceedings of the 28th International Conference on Machine Learning Icml 2011 · October 7, 2011 A convolutional factor-analysis model is developed, with the number of filters (factors) inferred via the beta process (BP) and hierarchical BP, for single-task and multi-task learning, respectively. The computation of the model parameters is implemented w ... Cite

Variational inference for stick-breaking beta process priors

Journal article Proceedings of the 28th International Conference on Machine Learning Icml 2011 · October 7, 2011 We present a variational Bayesian inference algorithm for the stick-breaking construction of the beta process. We derive an alternate representation of the beta process that is amenable to variational inference, and present a bound relating the truncated b ... Cite

Nonparametric Bayesian factor analysis of multiple time series

Journal article IEEE Workshop on Statistical Signal Processing Proceedings · September 5, 2011 We propose a nonparametric Bayesian factor analysis framework for characterization of multiple time-series. The proposed model automatically infers the number of factors and the noise/residual variance, and it is also able to cluster time series which beha ... Full text Cite

Dynamic relational topic model for social network analysis with noisy links

Journal article IEEE Workshop on Statistical Signal Processing Proceedings · September 5, 2011 A probabilistic framework is presented for joint analysis of text and links between nodes (e.g., people) in a time-evolving social network. Unlike existing approaches, the proposed model is able to handle noisy links, i.e., observed links between nodes for ... Full text Cite

Separating background and foregroundin video based on a nonparametric Bayesian model

Journal article IEEE Workshop on Statistical Signal Processing Proceedings · September 5, 2011 Separating background and foreground in video is a fundamental problem in computer vision. We present a Bayesian hierarchical model to address this challenge, and apply it to video with dynamic scenes. The model uses a nonparametric prior, a beta-bernoulli ... Full text Cite

Time-evolving modeling of social networks

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 18, 2011 A statistical framework for modeling and prediction of binary matrices is presented. The method is applied to social network analysis, specifically the database of US Supreme Court rulings. It is shown that the ruling behavior of Supreme Court judges can b ... Full text Cite

Nonparametric Bayesian feature selection for multi-task learning

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 18, 2011 We present a nonparametric Bayesian model for multi-task learning, with a focus on feature selection in binary classification. The model jointly identifies groups of similar tasks and selects the subset of features relevant to the tasks within each group. ... Full text Cite

Bayesian topic models for describing computer network behaviors

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 18, 2011 We consider the use of Bayesian topic models in the analysis of computer network traffic. Our approach utilizes latent Dirichlet allocation and time-varying dynamic latent Dirichlet allocation, with the goal of identifying significant co-occurrences of typ ... Full text Cite

Joint dictionary learning and topic modeling for image clustering

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 18, 2011 A new Bayesian model is proposed, integrating dictionary learning and topic modeling into a unified framework. The model is applied to cluster multiple images, and a subset of the images may be annotated. Example results are presented on the MNIST digit da ... Full text Cite

Covariate-dependent dictionary learning and sparse coding

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 18, 2011 A dependent hierarchical beta process (dHBP) is developed as a prior for data that may be represented in terms of a sparse set of latent features (dictionary elements), with covariate-dependent feature usage. The dHBP is applicable to general covariates an ... Full text Cite

Coherence, compressive sensing, and random sensor arrays

Journal article IEEE Antennas and Propagation Magazine · August 1, 2011 Random sensor arrays are examined from a compressive-sensing (CS) perspective, particularly in terms of the coherence of compressive-sensing matrices. It is demonstrated that the maximum sidelobe level of an array corresponds to the coherence of interest f ... Full text Cite

Temporal dynamics of host molecular responses differentiate symptomatic and asymptomatic influenza a infection.

Journal article PLoS Genet · August 2011 Exposure to influenza viruses is necessary, but not sufficient, for healthy human hosts to develop symptomatic illness. The host response is an important determinant of disease progression. In order to delineate host molecular responses that differentiate ... Full text Open Access Link to item Cite

Detection of viruses via statistical gene expression analysis.

Journal article IEEE Trans Biomed Eng · March 2011 We develop a new bayesian construction of the elastic net (ENet), with variational bayesian analysis. This modeling framework is motivated by analysis of gene expression data for viruses, with a focus on H3N2 and H1N1 influenza, as well as Rhino virus and ... Full text Link to item Cite

Foreword

Journal article European Urological Review · January 1, 2011 Cite

Predicting Viral Infection From High-Dimensional Biomarker Trajectories.

Journal article J Am Stat Assoc · January 1, 2011 There is often interest in predicting an individual's latent health status based on high-dimensional biomarkers that vary over time. Motivated by time-course gene expression array data that we have collected in two influenza challenge studies performed wit ... Full text Link to item Cite

Dependent hierarchical beta process for image interpolation and denoising

Journal article Journal of Machine Learning Research · January 1, 2011 A dependent hierarchical beta process (dHBP) is developed as a prior for data that may be represented in terms of a sparse set of latent features, with covariate-dependent feature usage. The dHBP is applicable to general covariates and data models, imposin ... Cite

Tree-Structured Infinite Sparse Factor Model.

Journal article Proceedings of the ... International Conference on Machine Learning. International Conference on Machine Learning · January 2011 A tree-structured multiplicative gamma process (TMGP) is developed, for inferring the depth of a tree-based factor-analysis model. This new model is coupled with the nested Chinese restaurant process, to nonparametrically infer the depth and width (structu ... Cite

Topic Modeling with Nonparametric Markov Tree.

Journal article Proceedings of the ... International Conference on Machine Learning. International Conference on Machine Learning · January 2011 A new hierarchical tree-based topic model is developed, based on nonparametric Bayesian techniques. The model has two unique attributes: (i) a child node in the tree may have more than one parent, with the goal of eliminating redundant sub-topics de ... Cite

Non-parametric Bayesian modeling and fusion of spatio-temporal information sources

Journal article Fusion 2011 14th International Conference on Information Fusion · January 1, 2011 We propose a Gaussian process (GP) factor analysis approach for modeling multiple spatio-temporal datasets with non-stationary spatial covariance structure. A novel kernel stick-breaking process based mixture of GPs is proposed to address the problem of no ... Cite

Logistic Stick-Breaking Process.

Journal article Journal of machine learning research : JMLR · January 2011 A logistic stick-breaking process (LSBP) is proposed for non-parametric clustering of general spatially- or temporally-dependent data, imposing the belief that proximate data are more likely to be clustered together. The sticks in the LSBP are realized via ... Cite

Nonparametric bayesian matrix completion

Journal article 2010 IEEE Sensor Array and Multichannel Signal Processing Workshop SAM 2010 · December 20, 2010 The Beta-Binomial processes are considered for inferring missing values in matrices. The model moves beyond the low-rank assumption, modeling the matrix columns as residing in a nonlinear subspace. Large-scale problems are considered via efficient Gibbs sa ... Full text Cite

Compressive Sensing on Manifolds Using a Nonparametric Mixture of Factor Analyzers: Algorithm and Performance Bounds.

Journal article IEEE transactions on signal processing : a publication of the IEEE Signal Processing Society · December 2010 Nonparametric Bayesian methods are employed to constitute a mixture of low-rank Gaussians, for data x ∈ ℝ N that are of high dimension N but are constrained to reside in a low-dimensional subregion of ℝ N< ... Full text Cite

Joint analysis of time-evolving binary matrices and associated documents

Journal article Advances in Neural Information Processing Systems 23: 24th Annual Conference on Neural Information Processing Systems 2010, NIPS 2010 · December 1, 2010 We consider problems for which one has incomplete binary matrices that evolve with time (e:g:, the votes of legislators on particular legislation, with each year characterized by a different such matrix). An objective of such analysis is to infer structure ... Cite

Discriminative sparse representations in hyperspectral imagery

Journal article Proceedings International Conference on Image Processing Icip · December 1, 2010 Recent advances in sparse modeling and dictionary learning for discriminative applications show high potential for numerous classification tasks. In this paper, we show that highly accurate material classification from hyperspectral imagery (HSI) can be ob ... Full text Cite

Nonparametric image interpolation and dictionary learning using spatially-dependent dirichlet and beta process priors

Journal article Proceedings International Conference on Image Processing Icip · December 1, 2010 We present a Bayesian model for image interpolation and dictionary learning that uses two nonparametric priors for sparse signal representations: the beta process and the Dirichlet process. Additionally, the model uses spatial information within the image ... Full text Cite

Tree-Structured compressive sensing with variational bayesian analysis

Journal article IEEE Signal Processing Letters · November 12, 2010 In compressive sensing (CS) the known structure in the transform coefficients may be leveraged to improve reconstruction accuracy. We here develop a hierarchical statistical model applicable to both wavelet and JPEG-based DCT bases, in which the tree struc ... Full text Cite

Bayesian inference of the number of factors in gene-expression analysis: application to human virus challenge studies.

Journal article BMC Bioinformatics · November 9, 2010 BACKGROUND: Nonparametric Bayesian techniques have been developed recently to extend the sophistication of factor models, allowing one to infer the number of appropriate factors from the observed data. We consider such techniques for sparse factor analysis ... Full text Open Access Link to item Cite

Sticky hidden Markov modeling of comparative genomic hybridization

Journal article IEEE Transactions on Signal Processing · October 1, 2010 We develop a sticky hidden Markov model (HMM) with a Dirichlet distribution (DD) prior, motivated by the problem of analyzing comparative genomic hybridization (CGH) data. As formulated the sticky DD-HMM prior is employed to infer the number of states in a ... Full text Cite

A stick-breaking construction of the beta process

Journal article Icml 2010 Proceedings 27th International Conference on Machine Learning · September 17, 2010 We present and derive a new stick-breaking construction of the beta process. The construction is closely related to a special case of the stick-breaking construction of the Dirich-let process (Sethuraman, 1994) applied to the beta distribution. We derive a ... Cite

Dynamic nonparametric bayesian models for analysis of music

Journal article Journal of the American Statistical Association · June 1, 2010 The dynamic hierarchical Dirichlet process (dHDP) is developed to model complex sequential data, with a focus on audio signals from music. The music is represented in terms of a sequence of discrete observations, and the sequence is modeled using a hidden ... Full text Open Access Cite

Hierarchical bayesian modeling of topics in time-stamped documents.

Journal article IEEE transactions on pattern analysis and machine intelligence · June 2010 We consider the problem of inferring and modeling topics in a sequence of documents with known publication dates. The documents at a given time are each characterized by a topic and the topics are drawn from a mixture model. The proposed model infers the c ... Full text Cite

Active learning and basis selection for kernel-based linear models: A bayesian perspective

Journal article IEEE Transactions on Signal Processing · May 1, 2010 We develop an active learning algorithm for kernel-based linear regression and classification. The proposed greedy algorithm employs a minimum-entropy criterion derived using a Bayesian interpretation of ridge regression. We assume access to a matrix, Φ∈ R ... Full text Cite

Classification with Incomplete Data Using Dirichlet Process Priors.

Journal article Journal of machine learning research : JMLR · March 2010 A non-parametric hierarchical Bayesian framework is developed for designing a classifier, based on a mixture of simple (linear) classifiers. Each simple classifier is termed a local "expert", and the number of experts and their construction are manifested ... Cite

Sparse signal recovery and acquisition with graphical models

Journal article IEEE Signal Processing Magazine · January 1, 2010 Many applications in digital signal processing, machine learning, and communications feature a linear regression problem in which unknown data points, hidden variables, or code words are projected into a lower dimensional space via © 2006 IEEE. ... Full text Cite

Sparse linear regression with beta process priors

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2010 A Bayesian approximation to finding the minimum ℓ0 norm solution for an underdetermined linear system is proposed that is based on the beta process prior. The beta process linear regression (BP-LR) model finds sparse solutions to the underdeterm ... Full text Cite

A nonparametric Bayesian model for kernel matrix completion

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2010 We present a nonparametric Bayesian model for completing low-rank, positive semidefinite matrices. Given an N x N matrix with underlying rank r, and noisy measured values and missing values with a symmetric pattern, the proposed Bayesian hierarchical model ... Full text Cite

Compressive particle filtering for target tracking

Journal article IEEE Workshop on Statistical Signal Processing Proceedings · December 25, 2009 This paper presents a novel compressive particle filter (henceforth CPF) for tracking one or more targets in video using a reduced set of observations. It is shown that, by applying compressive sensing ideas in a multi-particle-filter framework, it is poss ... Full text Cite

Nonparametric learning of dictionaries for sparse representation of sensor signals

Journal article Camsap 2009 2009 3rd IEEE International Workshop on Computational Advances in Multi Sensor Adaptive Processing · December 1, 2009 Nonparametric Bayesian techniques are considered for learning dictionaries for sparse data representations, with applications in sparse rendering of sensor data. The beta process is employed as a prior for learning the dictionary, and this non parametric m ... Full text Cite

Hidden markov models with stick-breaking priors

Journal article IEEE Transactions on Signal Processing · October 9, 2009 The number of states in a hidden Markov model (HMM) is an important parameter that has a critical impact on the inferred model. Bayesian approaches to addressing this issue include the nonparametric hierarchical Dirichlet process, which does not extend to ... Full text Cite

On the relationship between compressive sensing and random sensor arrays

Journal article IEEE Antennas and Propagation Magazine · October 1, 2009 Random sensor arrays are examined from a compressive-sensing (CS) perspective. It is demonstrated that the natural random-array projections manifested by the media Green's function are consistent with the projection-type measurements associated with compre ... Full text Cite

Exploiting signal sparseness for reduced-rate sampling

Journal article 2009 IEEE Long Island Systems Applications and Technology Conference Lisat 2009 · September 25, 2009 The rate at which signals are sampled in their native form (e.g. the "time domain" for many signals of interest) in order to capture all of the information of a signal - the so-called Nyquist rate in traditional sampling - equals one over twice the Fourier ... Full text Cite

Active learning for semi-supervised multi-task learning

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 23, 2009 We present an algorithm for active learning (adaptive selection of training data) within the context of semi-supervised multi-task classifier design. The semi-supervised multi-task classifier exploits manifold information provided by the unlabeled data, wh ... Full text Cite

Multi-task classification with infinite local experts

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 23, 2009 We propose a multi-task learning (MTL) framework for nonlinear classification, based on an infinite set of local experts in feature space. The usage of local experts enables sharing at the expert-level, encouraging the borrowing of information even if task ... Full text Cite

Dirichlet process mixture models with multiple modalities

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 23, 2009 The Dirichlet process can be used as a nonparametric prior for an infinite-dimensional probability mass function on the parameter space of a mixture model. The set of parameters over which it is defined is generally used for a single, parametric distributi ... Full text Cite

Music analysis with a Bayesian dynamic model

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 23, 2009 A Bayesian dynamic model is developed to model complex sequential data, with a focus on audio signals from music. The music is represented in terms of a sequence of discrete observations, and the sequence is modeled using a hidden Markov model (HMM) with t ... Full text Cite

Gene expression signatures diagnose influenza and other symptomatic respiratory viral infections in humans.

Journal article Cell Host Microbe · September 17, 2009 Acute respiratory infections (ARIs) are a common reason for seeking medical attention, and the threat of pandemic influenza will likely add to these numbers. Using human viral challenge studies with live rhinovirus, respiratory syncytial virus, and influen ... Full text Link to item Cite

Nonparametric factor analysis with beta process priors

Journal article ACM International Conference Proceeding Series · September 15, 2009 We propose a nonparametric extension to the factor analysis problem using a beta process prior. This beta process factor analysis (BPFA) model allows for a dataset to be decomposed into a linear combination of a sparse set of factors, providing information ... Full text Cite

Exploiting structure in wavelet-based bayesian compressive sensing

Journal article IEEE Transactions on Signal Processing · September 3, 2009 Bayesian compressive sensing (CS) is considered for signals and images that are sparse in a wavelet basis. The statistical structure of the wavelet coefficients is exploited explicitly in the proposed model, and, therefore, this framework goes beyond simpl ... Full text Cite

Semisupervised multitask learning.

Journal article IEEE transactions on pattern analysis and machine intelligence · June 2009 Context plays an important role when performing classification, and in this paper we examine context from two perspectives. First, the classification of items within a single task is placed within the context of distinct concurrent or previous classificati ... Full text Cite

Compressive sensing for multi-static scattering analysis

Journal article Journal of Computational Physics · May 20, 2009 Compressive sensing (CS) is a framework in which one attempts to measure a signal in a compressive mode, implying that fewer total measurements are required vis à vis direct sampling methods. Compressive sensing exploits the fact that the signal of interes ... Full text Cite

Migratory logistic regression for learning concept drift between two data sets with application to UXO sensing

Journal article IEEE Transactions on Geoscience and Remote Sensing · May 1, 2009 To achieve good generalization in supervised learning, the training and testing examples are usually required to be drawn from the same source distribution. In this paper, we propose a method to relax this requirement in the context of logistic regression. ... Full text Cite

Kernel-matching pursuits with arbitrary loss functions.

Journal article IEEE transactions on neural networks · March 2009 The purpose of this research is to develop a classifier capable of state-of-the-art performance in both computational efficiency and generalization ability while allowing the algorithm designer to choose arbitrary loss functions as appropriate for a give p ... Full text Cite

Semisupervised learning of hidden Markov models via a homotopy method.

Journal article IEEE transactions on pattern analysis and machine intelligence · February 2009 Hidden Markov model (HMM) classifier design is considered for the analysis of sequential data, incorporating both labeled and unlabeled data for training; the balance between the use of labeled and unlabeled data is controlled by an allocation parameter \l ... Full text Cite

Multitask compressive sensing

Journal article IEEE Transactions on Signal Processing · January 29, 2009 Compressive sensing (CS) is a framework whereby one performs N nonadaptive measurements to constitute a vector v∈ℝN with v used to recover an approximation u∈RℝM to a desired signal u∈RℝM with N≪ M; this is performed under ... Full text Cite

A Bayesian Model for Simultaneous Image Clustering, Annotation and Object Segmentation.

Journal article Advances in neural information processing systems · January 2009 A non-parametric Bayesian model is proposed for processing multiple images. The analysis employs image features and, when present, the words associated with accompanying annotations. The model clusters the images into classes, and each image is segmented i ... Cite

Learning to explore and exploit in POMDPs

Journal article Advances in Neural Information Processing Systems 22 Proceedings of the 2009 Conference · January 1, 2009 A fundamental objective in reinforcement learning is the maintenance of a proper balance between exploration and exploitation. This problem becomes more challenging when the agent can only partially observe the states of its environment. In this paper we p ... Cite

Non-parametric Bayesian dictionary learning for sparse image representations

Journal article Advances in Neural Information Processing Systems 22 Proceedings of the 2009 Conference · January 1, 2009 Non-parametric Bayesian techniques are considered for learning dictionaries for sparse image representations, with applications in denoising, inpainting and com-pressive sensing (CS). The beta process is employed as a prior for learning the dictionary, and ... Cite

Nonparametric factor analysis with beta process priors

Journal article Proceedings of the 26th International Conference on Machine Learning Icml 2009 · January 1, 2009 We propose a nonparametric extension to the factor analysis problem using a beta process prior. This beta process factor analysis (BP-FA) model allows for a dataset to be decomposed into a linear combination of a sparse set of factors, providing informatio ... Full text Cite

Multi-task reinforcement learning in partially observable stochastic environments

Journal article Journal of Machine Learning Research · January 1, 2009 We consider the problem of multi-task reinforcement learning (MTRL) in multiple partially observable stochastic environments. We introduce the regionalized policy representation (RPR) to characterize the agent's behavior in each environment. The RPR is a p ... Cite

On enhancing classification performance by exploiting multiple scattering

Journal article Applied Physics Letters · December 1, 2008 Using concepts developed in the fields of compressive sensing and random-projection-based embeddings, we consider classification of an object situated within a complex propagation environment. We demonstrate that propagation through such an environment may ... Full text Cite

Multitask classification by learning the task relevance

Journal article IEEE Signal Processing Letters · December 1, 2008 We consider the problem of multitask learning (MTL), in which we simultaneously learn classifiers for multiple data sets (tasks), with sharing of intertask data as appropriate. We introduce a set of relevance parameters that control the degree to which dat ... Full text Cite

Cybersecurity strategies: The QuERIES methodology

Journal article Computer · August 27, 2008 QuERIES offers a novel multidisciplinary approach to quantifying risk associated with security technologies resulting in investment-efficient cybersecurity strategies. © 2008 IEEE. ... Full text Cite

Multi-task learning for analyzing and sorting large databases of sequential data

Journal article IEEE Transactions on Signal Processing · August 1, 2008 A new hierarchical nonparametric Bayesian framework is proposed for the problem of multi-task learning (MTL) with sequential data. The models for multiple tasks, each characterized by sequential data, are learned jointly, and the intertask relationships ar ... Full text Cite

Bayesian compressive sensing

Journal article IEEE Transactions on Signal Processing · June 1, 2008 The data of interest are assumed to be represented as N-dimensional real vectors, and these vectors are compressible in some linear basis B, implying that the signal can be reconstructed accurately using only a small number M ≪ N of basis-function coeffici ... Full text Cite

The ATR center and ATRpedia

Journal article Proceedings of SPIE the International Society for Optical Engineering · May 20, 2008 The purpose of the Automatic Target Recognition (ATR) Center is to develop an environment conducive to producing theoretical and practical advances in the field of ATR. This will be accomplished by fostering intellectual growth of ATR practitioners at all ... Full text Cite

Infinite hidden Markov models for unusual-event detection in video.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · May 2008 We address the problem of unusual-event detection in a video sequence. Invariant subspace analysis (ISA) is used to extract features from the video, and the time-evolving properties of these features are modeled via an infinite hidden Markov model (iHMM), ... Full text Cite

An investigation of using the spectral characteristics from ground penetrating radar for landmine/clutter discrimination

Journal article IEEE Transactions on Geoscience and Remote Sensing · April 1, 2008 Ground penetrating radar (GPR)-based discrimination of landmines from clutter is known to be challenging due to the wide variability of possible clutter (e.g., rocks, roots, and general soil heterogeneity). This paper discusses the use of GPR frequency-dom ... Full text Cite

The matrix stick-breaking process: Flexible Bayes meta-analysis

Journal article Journal of the American Statistical Association · March 1, 2008 In analyzing data from multiple related studies, it often is of interest to borrow information across studies and to cluster similar studies. Although parametric hierarchical models are commonly used, of concern is sensitivity to the form chosen for the ra ... Full text Cite

In situ compressive sensing

Journal article Inverse Problems · February 1, 2008 Compressive sensing (CS) is a framework that exploits the compressible character of most natural signals, allowing the accurate measurement of an m-dimensional signal u in terms of n ≪ m measurements v. The CS measurements may be represented in terms of an ... Full text Cite

Broadband acoustic scattering measurements of underwater unexploded ordnance (UXO).

Journal article The Journal of the Acoustical Society of America · February 2008 In order to evaluate the potential for detection and identification of underwater unexploded ordnance (UXO) by exploiting their structural acoustic response, we carried out broadband monostatic scattering measurements over a full 360 degrees on UXO's (two ... Full text Cite

Detection of unexploded ordnance via efficient semisupervised and active learning

Journal article IEEE Transactions on Geoscience and Remote Sensing · January 1, 2008 Semisupervised learning and active learning are considered for unexploded ordnance (UXO) detection. Semisupervised learning algorithms are designed using both labeled and unlabeled data, where here labeled data correspond to sensor signatures for which the ... Full text Cite

Scattering from very large randomly rough surfaces using a Markov random field equivalent current

Journal article IEEE Transactions on Antennas and Propagation · January 1, 2008 A Markov random field (MRF) model is employed to learn the statistical properties of an equivalent current situated above a rough surface, where this equivalent current represents near-field electromagnetic scattered fields. The MRF parameters are learned ... Full text Cite

The dynamic hierarchical Dirichlet process

Journal article Proceedings of the 25th International Conference on Machine Learning · January 1, 2008 The dynamic hierarchical Dirichlet process (dHDP) is developed to model the time-evolving statistical properties of sequential data sets. The data collected at any time point are represented via a mixture associated with an appropriate underlying model, in ... Full text Cite

Hierarchical kernel stick-breaking process for multi-task image analysis

Journal article Proceedings of the 25th International Conference on Machine Learning · January 1, 2008 The kernel stick-breaking process (KSBP) is employed to segment general imagery, imposing the condition that patches (small blocks of pixels) that are spatially proximate are more likely to be associated with the same cluster (segment). The number of clust ... Full text Cite

Multi-task compressive sensing with dirichlet process priors

Journal article Proceedings of the 25th International Conference on Machine Learning · January 1, 2008 Compressive sensing (CS) is an emerging £eld that, under appropriate conditions, can signi£cantly reduce the number of measurements required for a given signal. In many applications, one is interested in multiple signals that may be measured in multiple CS ... Full text Cite

Experimental validation of a transport-based imaging method in highly scattering environments

Journal article Inverse Problems · December 1, 2007 We demonstrate the effectiveness of a transport-based reconstruction method for imaging in highly scattering environments. Experimentally measured wave energy data in the micro-wave regime are used to reconstruct extended inclusions buried in scattering me ... Full text Cite

Semi-supervised life-long learning with application to sensing

Journal article 2007 2nd IEEE International Workshop on Computational Advances in Multi Sensor Adaptive Processing Campsap · December 1, 2007 We present a semi-supervised multitask learning (MTL) framework, where we have multiple partially labeled data manifolds, each defining a classification task for which we wish to design a semi-supervised classifier. These different data sets may be observe ... Full text Cite

In situ compressive sensing

Journal article IEEE Workshop on Statistical Signal Processing Proceedings · December 1, 2007 Compressive sensing (CS) is a framework that exploits the compressible character of most natural signals, allowing the accurate measurement of an m-dimensional real signal u in terms of n≪m real measurements v. The CS measurements may be represented in ter ... Full text Cite

In situ compressive sensing

Journal article 2007 2nd IEEE International Workshop on Computational Advances in Multi Sensor Adaptive Processing Campsap · December 1, 2007 Compressive sensing (CS) is a framework that exploits the compressible character of most natural signals, allowing the accurate measurement of an m-dimensional real signal u in terms of n≪m real measurements v. The CS measurements may be represented in ter ... Full text Cite

Point-based policy iteration

Journal article Proceedings of the National Conference on Artificial Intelligence · November 28, 2007 We describe a point-based policy iteration (PBPI) algorithm for infinite-horizon POMDPs. PBPI replaces the exact policy improvement step of Hansen's policy iteration with point-based value iteration (PBVI). Despite being an approximate algorithm, PBPI is m ... Cite

Multi-task learning for underwater object classification

Journal article Proceedings of SPIE the International Society for Optical Engineering · November 15, 2007 The purpose of this research is to jointly learn multiple classification tasks by appropriately sharing information between similar tasks. In this setting, examples of different tasks include the discrimination of targets from non-targets by different sona ... Full text Cite

Music analysis using hidden Markov mixture models

Journal article IEEE Transactions on Signal Processing · November 1, 2007 We develop a hidden Markov mixture model based on a Dirichlet process (DP) prior, for representation of the statistics of sequential data for which a single hidden Markov model (HMM) may not be sufficient. The DP prior has an intrinsic clustering property ... Full text Cite

A bivariate gaussian model for unexploded ordnance classification with EMI data

Journal article IEEE Geoscience and Remote Sensing Letters · October 1, 2007 A bivariate Gaussian model is proposed for modeling spatially varying electromagnetic-induction (EMI) response of unexploded ordnance (UXO). This model is proposed for EMI sensors that do not exploit enough physics to warrant using the popular magnetic-dip ... Full text Cite

Quadratically gated mixture of experts for incomplete data classification

Journal article ACM International Conference Proceeding Series · August 23, 2007 We introduce quadratically gated mixture of experts (QGME), a statistical model for multi-class nonlinear classification. The QGME is formulated in the setting of incomplete data, where the data values are partially observed. We show that the missing value ... Full text Cite

Multi-task learning for sequential data via iHMMs and the nested Dirichlet process

Journal article ACM International Conference Proceeding Series · August 23, 2007 A new hierarchical nonparametric Bayesian model is proposed for the problem of multitask learning (MTL) with sequential data. Sequential data are typically modeled with a hidden Markov model (HMM), for which one often must choose an appropriate model struc ... Full text Cite

Bayesian compressive sensing and projection optimization

Journal article ACM International Conference Proceeding Series · August 23, 2007 This paper introduces a new problem for which machine-learning tools may make an impact. The problem considered is termed "compressive sensing", in which a real signal of dimension N is measured accurately based on K << N real measurements. This is achieve ... Full text Cite

The matrix stick-breaking process for flexible multi-task learning

Journal article ACM International Conference Proceeding Series · August 23, 2007 In multi-task learning our goal is to design regression or classification models for each of the tasks and appropriately share information between tasks. A Dirichlet process (DP) prior can be used to encourage task clustering. However, the DP prior does no ... Full text Cite

Wideband array imaging of a target situated in an unknown random media

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 6, 2007 We propose two new methods for Wideband array signal imaging for targets situated in unknown random media. First, a normalized coherent interferometric (N-CINT) imaging algorithm is developed based on coherent interferometric (CINT) imaging theory, yieldin ... Full text Cite

Learning classifiers on a partially labeled data manifold

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 6, 2007 We present an algorithm for learning parametric classifiers on a partially labeled data manifold, based on a graph representation of the manifold. The unlabeled data are utilized by basing classifier learning on neighborhoods, formed via Markov random, wal ... Full text Cite

Multi-aspect target classification and detection via the infinite hidden markov model

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 6, 2007 A new multi-aspect target detection method is presented based on the infinite hidden Markov model (iHMM). The scattering of waves from multiple targets is modeled as an iHMM with the number of underlying states treated as infinite, from which a full poster ... Full text Cite

Dirichlet process HMM mixture models with application to music analysis

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 6, 2007 A hidden Markov mixture model is developed using a Dirichlet process (DP) prior, to represent the statistics of sequential data for which a single hidden Markov model (HMM) may not be sufficient. The DP prior has an intrinsic clustering property that encou ... Full text Cite

Classification of unexploded ordnance using incomplete multisensor multiresolution data

Journal article Ieee Transactions On Geoscience And Remote Sensing · July 2007 We address the problem of unexploded ordnance (UXO) detection in which data to be classified are available from multiple sensor modalities and multiple resolutions. Specifically, features are extracted from measured magnetometer and electromagnetic inducti ... Cite

Volumetric fast multipole method for modeling Schrödinger's equation

Journal article Journal of Computational Physics · June 10, 2007 A volume integral equation method is presented for solving Schrödinger's equation for three-dimensional quantum structures. The method is applicable to problems with arbitrary geometry and potential distribution, with unknowns required only in the part of ... Full text Cite

Adaptive multimodality sensing of landmines

Journal article IEEE Transactions on Geoscience and Remote Sensing · June 1, 2007 The problem of adaptive multimodality sensing of landmines is considered based on electromagnetic induction (EMI) and ground-penetrating radar (GPR) sensors. Two formulations are considered based on a partially observable Markov decision process (POMDP) fr ... Full text Cite

Nonmyopic multiaspect sensing with partially observable Markov decision processes

Journal article IEEE Transactions on Signal Processing · June 1, 2007 We consider the problem of sensing a concealed or distant target by interrogation from multiple sensors situated on a single platform. The available actions that may be taken are selection of the next relative target-platform orientation and the next senso ... Full text Cite

Three-dimensional Bayesian inversion with application to subsurface sensing

Journal article IEEE Transactions on Geoscience and Remote Sensing · May 1, 2007 A Bayesian formalism is considered for inverting for the parameters of a heterogeneity profile based on measured scattering data. It is shown that the typical use of regularization (e.g., Thikonov) corresponds to a maximum a posteriori point approximation ... Full text Cite

Multiaspect target detection via the infinite hidden Markov model.

Journal article The Journal of the Acoustical Society of America · May 2007 A new multiaspect target detection method is presented based on the infinite hidden Markov model (iHMM). The scattering of waves from a target is modeled as an iHMM with the number of underlying states treated as infinite, from which a full posterior distr ... Full text Cite

Cost-sensitive feature acquisition and classification

Journal article Pattern Recognition · May 1, 2007 There are many sensing challenges for which one must balance the effectiveness of a given measurement with the associated sensing cost. For example, when performing a diagnosis a doctor must balance the cost and benefit of a given test (measurement), and t ... Full text Cite

Electromagnetic Target Detection in Uncertain Media: Time-Reversal and Minimum-Variance Algorithms

Journal article IEEE Transactions on Geoscience and Remote Sensing · April 7, 2007 An experimental study is performed on imaging targets that are situated in a highly scattering environment, employing electromagnetic time-reversal methods. A particular focus is placed on performance when the electrical properties of the background enviro ... Full text Cite

Active learning applied to RCS computations with nonuniform sampling using different objective functions

Journal article IEEE Transactions on Antennas and Propagation · April 1, 2007 An active learning framework is introduced to reduce the number of frequencies and angles one must consider for wideband monostatic scattering computations or measurements. This method is used to optimally select those frequencies and angles that would be ... Full text Cite

On classification with incomplete data.

Journal article IEEE transactions on pattern analysis and machine intelligence · March 2007 We address the incomplete-data problem in which feature vectors to be classified are missing data (features). A (supervised) logistic regression algorithm for the classification of incomplete data is developed. Single or multiple imputation for the missing ... Full text Cite

Electromagnetic time-reversal source localization in changing media: Experiment and analysis

Journal article IEEE Transactions on Antennas and Propagation · February 1, 2007 An experimental study is performed on electromagnetic time reversal in highly scattering environments, with a particular focus on performance when environmental conditions change. In particular, we consider the case for which there is a mismatch between th ... Full text Cite

Multi-task learning for classification with Dirichlet process priors

Journal article Journal of Machine Learning Research · January 1, 2007 Consider the problem of learning logistic-regression models for multiple classification tasks, where the training data set for each task is not drawn from the same statistical distribution. In such a multi-task learning (MTL) scenario, it is necessary to i ... Cite

Infinite hidden Markov models and ISA features for unusual-event detection in video

Journal article Proceedings International Conference on Image Processing Icip · January 1, 2007 We address the problem of unusual-event detection in a video sequence. Invariant subspace analysis (ISA) is used to extract features from the video, and the time-evolving properties of these features are modeled via an infinite hidden Markov model (iHMM), ... Full text Cite

A kernel machine framework for feature optimization in multi-frequency sonar imagery

Journal article Oceans 2006 · December 1, 2006 The purpose of this research is to optimize the extraction of classification features. This includes the optimal adjustment of parameters used to compute features as well as an objective and quantitative method to assist in choosing a priori data collectio ... Full text Cite

Region-based value iteration for partially observable Markov decision processes

Journal article ACM International Conference Proceeding Series · December 1, 2006 An approximate region-based value iteration (RBVI) algorithm is proposed to find the optimal policy for a partially observable Markov decision process (POMDP). The proposed RBVI approximates the true polyhedral partition of the belief simplex with an ellip ... Full text Cite

A reward-directed Bayesian classifier

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · December 1, 2006 We consider a classification problem wherein the class features are not given a priori. The classifier is responsible for selecting the features, to minimize the cost of observing features while also maximizing the classification performance. We propose a ... Cite

Wave-based signal processing

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · December 1, 2006 In many ways the electromagnetics and signal processing communities are at similar levels of development; in both fields there are many mature techniques available to address problems of interest. It is a good time to cross-fertilize between these fields. ... Full text Cite

Homotopy-based semi-supervised hidden Markov tree for texture analysis

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · December 1, 2006 A semi-supervised hidden Markov tree (HMT) model is developed for texture analysis, incorporating both labeled and unlabeled data for training; the optimal balance between labeled and unlabeled data is estimated via the homotopy method. In traditional EM-b ... Cite

Incremental least squares policy iteration for POMDPs

Journal article Proceedings of the National Conference on Artificial Intelligence · November 13, 2006 We present a new algorithm, called incremental least squares policy iteration (ILSPI), for finding the infinite-horizon stationary policy for partially observable Markov decision processes (POMDPs). The ILSPI algorithm computes a basis representation of th ... Cite

Region-based value iteration for partially observable Markov decision processes

Journal article Icml 2006 Proceedings of the 23rd International Conference on Machine Learning · October 6, 2006 An approximate region-based value iteration (RBVI) algorithm is proposed to find the optimal policy for a partially observable Markov decision process (POMDP). The proposed RBVI approximates the true polyhedral partition of the belief simplex with an ellip ... Cite

Pseudospectral method based on prolate spheroidal wave functions for semiconductor nanodevice simulation

Journal article Computer Physics Communications · July 15, 2006 We solve Schrödinger's equation for semiconductor nanodevices by applying prolate spheroidal wave functions of order zero as basis functions in the pseudospectral method. When the functions involved in the problem are bandlimited, the prolate pseudospectra ... Full text Cite

Texture analysis with variational hidden Markov trees

Journal article IEEE Transactions on Signal Processing · June 1, 2006 A variational Bayes formulation of the hidden Markov tree (HMT) model is proposed for texture analysis, utilizing a multilevel wavelet decomposition of imagery. The variational method yields an approximation to the full posterior of the HMT parameters. Tex ... Full text Cite

Variational Bayes for continuous hidden Markov models and its application to active learning.

Journal article IEEE transactions on pattern analysis and machine intelligence · April 2006 In this paper, we present a varitional Bayes (VB) framework for learning continuous hidden Markov models (CHMMs), and we examine the VB framework within active learning. Unlike a maximum likelihood or maximum a posteriori training procedure, which yield a ... Full text Cite

A modified SPIHT algorithm for image coding with a joint MSE and classification distortion measure.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · March 2006 The set partitioning in hierarchical trees (SPIHT) algorithm is an efficient wavelet-based progressive image-compression technique, designed to minimize the mean-squared error (MSE) between the original and decoded imagery. However, the MSE-based distortio ... Full text Cite

Rapid prolate pseudospectral differentiation and interpolation with the fast multipole method

Journal article SIAM Journal on Scientific Computing · January 1, 2006 Pseudospectral methods utilizing prolate spheroidal wave functions as basis functions have been shown to possess advantages over the conventional pseudospectral methods based on trigonometric and orthogonal polynomials. However, the spectral differentiatio ... Full text Cite

Double-sided exponentially tapered GPR antenna and its transmission line feed structure

Journal article IEEE Transactions on Antennas and Propagation · 2006 A double-sided broadband antenna for applications including ground-penetrating radar for detecting buried target is described. When compared with traditional coplanar-strip antennas, a better performance is achieved with a more practical design for constru ... Full text Cite

Pseudospectral method based on prolate spheroidal wave functions for frequency-domain electromagnetic simulations

Journal article IEEE Transactions on Antennas and Propagation · December 1, 2005 We apply prolate spheroidal wave functions of order zero as basis functions in the pseudospectral method for frequency-domain electromagnetic simulation problems. Like the traditional pseudospectral frequency-domain (PSFD) methods based on Chebyshev and Le ... Full text Cite

Radial basis function network for multi-task learning

Journal article Advances in Neural Information Processing Systems · December 1, 2005 We extend radial basis function (RBF) networks to the scenario in which multiple correlated tasks are learned simultaneously, and present the corresponding learning algorithms. We develop the algorithms for learning the network structure, in either a super ... Cite

Incomplete-data classification using logistic regression

Journal article Icml 2005 Proceedings of the 22nd International Conference on Machine Learning · December 1, 2005 A logistic regression classification algorithm is developed for problems in which the feature vectors may be missing data (features). Single or multiple imputation for the missing data is avoided by performing analytic integration with an estimated conditi ... Cite

Direct algorithm for computation of derivatives of the Daubechies basis functions

Journal article Applied Mathematics and Computation · November 15, 2005 We extend the direct algorithm for computing the derivatives of the compactly supported Daubechies N-vanishing-moment basis functions. The method yields exact values at dyadic rationals for the nth derivative (0 ≤ n ≤ N - 1) of the basis functions, when it ... Full text Cite

Image technique for multiresolution time-domain using nonsymmetric basis functions

Journal article Microwave and Optical Technology Letters · October 5, 2005 The image technique is required in order to model perfect-electric- conductor and perfect-magnetic-conductor boundary conditions using the multiresolution time-domain (MRTD) method. The strategy of directly imposing the image principle on the field-expansi ... Full text Cite

Adaptive multiaspect target classification and detection with hidden Markov models

Journal article IEEE Sensors Journal · October 1, 2005 Target detection and classification are considered based on backscattered signals observed from a sequence of target-sensor orientations, with the measurements performed as a function of orientation (angle) at a fixed range. The theory of optimal experimen ... Full text Cite

Nonuniform frequency sampling with active learning: Application to wide-band frequency-domain modeling and design

Journal article IEEE Transactions on Antennas and Propagation · September 1, 2005 One must employ many frequency points to synthesize a wide-band time-domain signal scattered or radiated from a given linear device. If the structure is large relative to wavelengths of interest, the large number of required frequency-domain computations m ... Full text Cite

Ridgelet-based implementation of multiresolution time domain

Journal article IEEE Transactions on Antennas and Propagation · August 1, 2005 The theory of ridgelet-based analysis of time-domain wave propagation and scattering is developed. Some of the advantages of using ridgelets as compared to conventional wavelets are as follows. First, ridgelets often require less expansion coefficients (un ... Full text Cite

Rate-distortion bound for joint compression and classification with application to multiaspect scattering

Journal article IEEE Sensors Journal · June 1, 2005 Rate-distortion analysis is applied to the problem of joint compression and classification. A Lagrangian distortion measure is used to consider both the Euclidean error in reconstructing the original data as well as the classification performance. The boun ... Full text Cite

Sparse multinomial logistic regression: fast algorithms and generalization bounds.

Journal article IEEE transactions on pattern analysis and machine intelligence · June 2005 Recently developed methods for learning sparse classifiers are among the state-of-the-art in supervised learning. These methods learn classifiers that incorporate weighted sums of basis functions with sparsity-promoting priors encouraging the weight estima ... Full text Cite

Analysis of scattering from very large three-dimensional rough surfaces using MLFMM and ray-based analyses

Journal article IEEE Antennas and Propagation Magazine · June 1, 2005 Several techniques are considered for the analysis of electromagnetic scattering from rough ocean surfaces. A rigorous Multi-Level Fast Multipole Method (MLFMM) is employed, as well as a high-frequency ray-based solution. The MLFMM analysis is implemented ... Full text Cite

Order of accuracy analysis for multiresolution time-domain using daubechies bases

Journal article Microwave and Optical Technology Letters · May 20, 2005 In this paper, the spatial order of accuracy of multiresolution time-domain methods using basis functions from the Daubechies family are studied. It is observed that MRTD methods using scaling functions from the Daubechies N-vanishing-moment orthonormal fa ... Full text Cite

Time-reversal imaging for classification of submerged elastic targets via Gibbs sampling and the Relevance Vector Machine.

Journal article The Journal of the Acoustical Society of America · April 2005 Time-reversal imaging (TRI) is analogous to matched-field processing, although TRI is typically very wideband and is appropriate for subsequent target classification (in addition to localization). Time-reversal techniques, as applied to acoustic target cla ... Full text Cite

Active learning for detection of mine-like objects in side-scan sonar imagery

Journal article IEEE Journal of Oceanic Engineering · April 1, 2005 A data-adaptive algorithm is presented for the selection of the basis functions and training data used in classifier design with application to sensing mine-like targets with a side-scan sonar. Automatic detection of mine-like targets using side-scan sonar ... Full text Cite

Electromagnetic time-reversal imaging of a target in a cluttered environment

Journal article IEEE Transactions on Antennas and Propagation · 2005 Electromagnetic time-reversal imaging is addressed for a target situated in a cluttered background. We first investigate the theory of electromagnetic time-reversal imaging, followed by an experimental demonstration. A transmitter-receiver antenna array is ... Full text Cite

Analysis, design, and construction of a broadband balun for coaxial-to-planar transmission lines

Journal article Microwave and Optical Technology Letters · 2005 For microwave signals within the 0.5 to 10.5 GHz range, we present an integrated broadband balun that carries signals from co-axial cable to microstrip to parallel-plate line. We use this arrangement as the feed line for a broadband double-sided antenna, a ... Full text Cite

A bayesian approach to unsupervised feature selection and density estimation using expectation propagation

Journal article Proceedings 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Cvpr 2005 · January 1, 2005 We propose an approximate Bayesian approach for unsupervised feature selection and density estimation, where the importance of the features for clustering is used as the measure for feature selection. Traditional maximum-likelihood (ML) model-parameter opt ... Full text Cite

Logistic regression with an auxiliary data source

Journal article Icml 2005 Proceedings of the 22nd International Conference on Machine Learning · January 1, 2005 To achieve good generalization in supervised learning, the training and testing examples are usually required to be drawn from the same source distribution. In this paper we propose a method to relax this requirement in the context of logistic regression. ... Full text Cite

Wideband frequency response of low metal mines

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 20, 2004 Extensive studies of in-air testing of various metal detectors have been previously performed for a wide variety of targets and operating conditions. 1,2 Using similar targets, we conducted a preliminary evaluation of a laboratory prototype wide ... Full text Cite

Inverse scattering with sparse Bayesian vector regression

Journal article Inverse Problems · December 1, 2004 A Bayesian formulation is employed to develop a sparse vector regression model. The model is used to characterize the connection between measured vector scattered-field data x and the underlying target responsible for these fields, characterized by the par ... Full text Cite

Detection of buried targets via active selection of labeled data: Application to sensing subsurface UXO

Journal article IEEE Transactions on Geoscience and Remote Sensing · November 1, 2004 When sensing subsurface targets, such as landmines and unexploded ordnance (UXO), the target signatures are typically a strong function of environmental and historical circumstances. Consequently, it is difficult to constitute a universal training set for ... Full text Cite

Adaptive multi-aspect target classification and detection with hidden Markov models

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · October 7, 2004 We consider target classification and detection based on back-scattered observations measured from a sequence of target-sensor orientations. The multi-aspect scattered waves from a given target are modeled with a hidden Markov model (HMM). The targets are ... Cite

Time-reversal imaging and classification for distant targets in a shallow water channel

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · October 7, 2004 Time-reversal imaging (TRI) is analogous to matched-field processing, although TRI is typically very wideband and is capable of performing target classification (in addition to localization). In this paper we apply the time-reversal technique to locate man ... Cite

Kernel matching pursuits prioritization of wavelet coefficients for SPIHT image coding

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 28, 2004 The set partitioning in hierarchical trees (SPIHT), an efficient wavelet-based progressive image-compression scheme, is oriented to minimize the mean-squared error (MSE) between the original and decoded imagery. In this paper, we use the kernel matching pu ... Cite

A Bayesian approach to joint feature selection and classifier design.

Journal article IEEE transactions on pattern analysis and machine intelligence · September 2004 This paper adopts a Bayesian approach to simultaneously learn both an optimal nonlinear classifier and a subset of predictor variables (or features) that are most relevant to the classification task. The approach uses heavy-tailed priors to promote sparsit ... Full text Cite

Application of the theory of optimal experiments to adaptive electromagnetic-induction sensing of buried targets.

Journal article IEEE transactions on pattern analysis and machine intelligence · August 2004 A mobile electromagnetic-induction (EMI) sensor is considered for detection and characterization of buried conducting and/or ferrous targets. The sensor may be placed on a robot and, here, we consider design of an optimal adaptive-search strategy. A freque ... Full text Cite

Analysis of wideband forward looking synthetic aperture radar for sensing land mines

Journal article Radio Science · July 1, 2004 Signal processing algorithms are considered for the analysis of wideband, forward looking synthetic aperture radar data and for sensing metal and plastic land mines, with principal application to unpaved roads. Simple prescreening algorithms are considered ... Full text Cite

Application of the biorthogonal multiresolution time-domain method to the analysis of elastic-wave interactions with buried targets

Journal article IEEE Transactions on Geoscience and Remote Sensing · July 1, 2004 The biorthogonal multiresolution time-domain (Bi-MRTD) method is introduced for the analysis of elastic-wave interaction with buried targets. We provide a detailed discussion on implementation of the perfectly matched layer and on treatment of the interfac ... Full text Cite

Three-dimensional inverse scattering of a dielectric target embedded in a lossy half-space

Journal article IEEE Transactions on Geoscience and Remote Sensing · May 1, 2004 A modified iterative Born method is applied for three-dimensional inversion of a lossless dielectric target embedded in a lossy half-space. The forward solver employs a modified form of the extended Born method, and the half-space Green's function is compu ... Full text Cite

Classification of distant targets situated near channel bottoms

Journal article Journal of the Acoustical Society of America · March 1, 2004 Identification algorithms are considered for a class of targets situated near the bottom of a water channel. It is assumed that the target-sensor distance relative to the channel depth is such that a ray-based representation of the scattered fields is appr ... Full text Cite

Target identification from multi-aspect high range-resolution radar signatures using a hidden Markov model

Journal article IEICE Transactions on Electronics · January 1, 2004 Identification of targets using sequential high range-resolution (HRR) radar signatures is studied. Classifiers are designed by using hidden Markov models (HMMs) to characterize the sequential information, in multi-aspect HRR signatures. The higher-order m ... Cite

Volumetric MLFMA formulation for dielectric targets in the presence of a half-space

Journal article Radio Science · 2004 The fast multipole method (FMM) and the multilevel fast multipole algorithm (MLFMA) are extended to the analysis of volumetric electric field integral equations (VEFIE) for targets in the presence of a half-space, to calculate the electromagnetic fields sc ... Cite

Multilevel fast multipole calibration of ray models with application to wireless propagation

Journal article IEEE Transactions on Antennas and Propagation · 2004 Ray tracing models are known to yield accurate results if a sufficient number of terms (e.g. diffraction mechanisms) are accounted for in the asymptotic formulation. For wireless applications one desires a ray analysis in which the fewest number of terms a ... Full text Link to item Cite

Wideband time-reversal imaging of an elastic target in an acoustic waveguide.

Journal article The Journal of the Acoustical Society of America · January 2004 Time-reversal is addressed for imaging elastic targets situated in an acoustic waveguide. It is assumed that the target-sensor range is large relative to the channel depth. We investigate the theory of wideband time-reversal imaging of an extended elastic ... Full text Cite

Airport detection in large aerial optical imagery

Journal article ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings · 2004 A method to detect airports in large aerial optical imagery is considered. Combining texture segmentation and shape detection, this method shows advantages in analyzing large aerial imagery. First, large aerial images are segmented and interpreted accordin ... Cite

Active selection of labeled data for target detection

Journal article ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings · 2004 An information-theoretic approach is developed for target detection, with active selection of training set, directly from the site-specific measured data For the proposed kernel-based algorithm, a set of basis functions are defined first to characterize th ... Cite

Joint classifier and feature optimization for comprehensive cancer diagnosis using gene expression data.

Journal article Journal of computational biology : a journal of computational molecular cell biology · January 2004 Recent research has demonstrated quite convincingly that accurate cancer diagnosis can be achieved by constructing classifiers that are designed to compare the gene expression profile of a tissue of unknown cancer status to a database of stored expression ... Full text Cite

Time-domain target detection using a double-sided broadband antenna

Journal article IEEE Antennas and Propagation Society, AP-S International Symposium (Digest) · 2004 The application of a double-sided broadband antenna for time-domain target detection was discussed. A mathematical model to simulate the structure and performance of the antenna was established, based on dipole antenna array theory. A formula for the elect ... Cite

Quantization of Multiaspect Scattering Data: Target Classification and Pose Estimation

Journal article IEEE Transactions on Signal Processing · December 1, 2003 In many sensing scenarios, the observed scattered waveforms must be quantized for subsequent transmission over a communication channel. Rate-distortion theory plays an important role in defining the bit rate required to achieve a desired distortion. The di ... Full text Cite

Time-reversal imaging of distant targets in a shallow water channel

Journal article Oceans Conference Record IEEE · December 1, 2003 Time-reversal imaging (TRI) is analogous to matched-field processing, although TRI is typically very wideband and it affords the potential of target classification (in addition to localization). In this paper we apply TRI to mine-like targets situated in s ... Cite

Relevance vector machine feature selection and classification for underwater targets

Journal article Oceans Conference Record IEEE · December 1, 2003 Feature selection is an important issue in detection and classification of underwater targets. Often feature selection is performed only indirectly linked to the ultimate objective: target classification. In this paper we consider several techniques for fe ... Cite

Broadband frequency-domain magnetic system for landmine/UXO detection and discrimination

Journal article Proceedings of SPIE the International Society for Optical Engineering · November 26, 2003 Using broadband magnetoresistive sensors, Quantum Magnetics is developing a metal detector for landmine/UXO detection and discrimination. When completed, this active system will be incorporated into a passive man-portable gradiometer system being developed ... Full text Cite

Analysis of the CDF biorthogonal MRTD method with application to PEC targets

Journal article IEEE Transactions on Microwave Theory and Techniques · September 1, 2003 We consider the biorthogonal Cohen-Daubechies-Feauveau (CDF) wavelet family in the context of a biorthogonal multiresolution time-domain (bi-MRTD) analysis. A disadvantage of previous bi-MRTD analyses is an inability to handle abrupt changes in material pr ... Full text Cite

MLFMA-based quasi-direct analysis of scattering from electrically large targets

Journal article IEEE Transactions on Antennas and Propagation · August 1, 2003 The multilevel fast multipole algorithm (MLFMA) is traditionally employed in the context of an iterative matrix solver, in which the MLFMA is utilized to implement the underlying matrix product with N log N complexity, where N represents the number of unkn ... Full text Cite

Rate-distortion analysis of discrete-HMM pose estimation via multiaspect scattering data

Journal article IEEE Transactions on Pattern Analysis and Machine Intelligence · July 1, 2003 We consider the problem of estimating the pose of a target based on a sequence of scattered waveforms measured at multiple target-sensor orientations. Using a hidden Markov model (HMM) representation of the scattered-waveform sequence, pose estimation redu ... Full text Cite

Sensing of unexploded ordnance with magnetometer and induction data: Theory and signal processing

Journal article IEEE Transactions on Geoscience and Remote Sensing · May 1, 2003 We consider the detection of subsurface unexploded ordnance via magnetometer and electromagnetic-induction (EMI) sensors. Detection performance is presented, using model-based signal processing algorithms. We first develop and validate the parametric model ... Full text Cite

Three-dimensional biorthogonal multiresolution time-domain method and its application to electromagnetic scattering problems

Journal article IEEE Transactions on Antennas and Propagation · May 1, 2003 A three-dimensional (3-D) multiresolutlon time-domain (MRTD) analysis is presented based on a biorthogonal-wavelet expansion, with application to electromagnetic-scattering problems. We employ the Cohen-Daubechies-Feauveau (CDF) biorthogonal wavelet basis, ... Full text Cite

Parallel implementation of the biorthogonal multiresolution time-domain method.

Journal article Journal of the Optical Society of America. A, Optics, image science, and vision · May 2003 The three-dimensional biorthogonal multiresolution time-domain (Bi-MRTD) method is presented for both free-space and half-space scattering problems. The perfectly matched layer (PML) is used as an absorbing boundary condition. It has been shown that improv ... Full text Cite

Physics-based detection of targets in SAR imagery usingsupport vector machines

Journal article IEEE Sensors Journal · April 1, 2003 Radar scattering from an illuminated object is often highly dependent on the target-sensor orientation. In conjunction with physics based feature extraction, the exploitation of aspect-dependent information has led to successful improvements in the detecti ... Full text Cite

Class-based target identification with multiaspect scattering data

Journal article IEEE Journal of Oceanic Engineering · April 1, 2003 In underwater sensing applications, it is often difficult to train a classifier in advance for all targets that may be seen during testing, due to the large number of targets that may be encountered. We therefore partition the training data into target cla ... Full text Cite

MLFMA analysis of scattering from multiple targets in the presence of a half-space

Journal article IEEE Transactions on Antennas and Propagation · 2003 The multilevel fast multipole algorithm (MLFMA) is applied to the analysis of plane-wave scattering from multiple conducting and/or dielectric targets, of arbitrary shape, situated in the presence of a dielectric half-space. The multiple-target scattering ... Full text Link to item Cite

Well-Conditioned MLFMA Formulation for Closed PEC Targets in the Vicinity of a Half Space

Journal article IEEE Transactions on Antennas and Propagation · 2003 The multilevel fast multipole algorithm (MLFMA) is applied to the problem of scattering from a closed perfect electric conductor (PEC) in the presence of a half space. The combined-field integral equation (CFIE) employs a new electric-field integral equati ... Full text Link to item Cite

Unexploded ordnance detection using Bayesian physics-based data fusion

Journal article Integrated Computer Aided Engineering · January 1, 2003 Detection of unexploded ordnance (UXO) represents a major challenge on closed, closing, and transferred military ranges as well as on active installations. On sites contaminated with UXO, extensive surface and sub-surface clutter is also present. Tradition ... Full text Cite

Scalable multilevel fast multipole method for multiple targets in the vicinity of a half space

Journal article IEEE Transactions on Geoscience and Remote Sensing · 2003 We extend the multilevel fast multipole algorithm (MLFMA) to the case of electromagnetic scattering from an arbitrary number of dielectric and/or perfectly conducting targets in the presence of a half space. This multitarget MLFMA is implemented in an iter ... Full text Link to item Cite

Time-Reversal Imaging for Wideband Underwater Target Classification

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2003 Time-reversal imaging is addressed for sensing an elastic target situated in an acoustic waveguide. It is demonstrated that the channel parameters associated with a given measurement may be determined via a genetic-algorithm (GA) search in parameter space. ... Cite

ICA with multiple quadratic constraints

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2003 The independent component analysis (ICA) with a single quadratic constraint on each source signal or column of the mixing matrix is extended to the case of multiple quadratic constraints. The criterion of Joint Approximate Diagonalization of Eigen-matrices ... Cite

Context-based graphical modeling for wavelet domain signal processing

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2003 Wavelet-domain hidden Markov tree (HMT) modeling provides a powerful approach to capture the underlying statistics of the wavelet coefficients. We develop a mutual information-based information-theoretic approach to quantify the interactions between the wa ... Cite

Sequential modeling for identifying CpG island locations in human genome

Journal article IEEE Signal Processing Letters · December 1, 2002 We consider several sequential processing algorithms for identifying genes in human DNA, based on detecting CpG ("C proceeds G") islands. The algorithms are designed to capture the underlying statistical structure in a DNA sequence. Sequential processing u ... Full text Cite

Support vector machines for improved multiaspect target recognition using the fisher kernel scores of hidden Markov models

Journal article Proceedings of the Joint Conference on Information Sciences · December 1, 2002 In conjunction with physics-based feature extraction, Hidden Markov Model (HMM) classifiers have been used successfully to fuse scattering data from multiple target orientations where the target-sensor orientation is generally unknown or "hidden". The use ... Cite

Infrared-image classification using hidden Markov trees

Journal article IEEE Transactions on Pattern Analysis and Machine Intelligence · October 1, 2002 An image of a three-dimensional target is generally characterized by the visible target subcomponents, with these dictated by the target-sensor orientation (target pose). An image often changes quickly with variable pose. We define a class as a set of cont ... Full text Cite

Identification of ground targets from sequential high-range-resolution radar signatures

Journal article IEEE Transactions on Aerospace and Electronic Systems · October 1, 2002 An approach to identifying targets from sequential high-range-resolution (HRR) radar signatures is presented. In particular, a hidden Markov model (HMM) is employed to characterize the sequential information contained in multiaspect HRR target signatures. ... Full text Cite

Scattering analysis by the multiresolution time-domain method using compactly supported wavelet systems

Journal article IEEE Transactions on Microwave Theory and Techniques · July 1, 2002 We present a formulation of the multiresolution time-domain (MRTD) algorithm using scaling and one-level wavelet basis functions, for orthonormal Daubechies and biorthogonal Cohen-Daubechies-Feauveau (CDF) wavelet families. We address the issue of the anal ... Full text Cite

Application of Haar-wavelet-based multiresolution time-domain schemes to electromagnetic scattering problems

Journal article IEEE Transactions on Antennas and Propagation · June 1, 2002 The multiresolution time-domain (MRTD) algorithm is applied to the problem of general two-dimensional electromagnetic scattering. A Haar wavelet expansion is utilized. A parallel between Haar MRTD and the classic Yee finite-difference time-domain (FDTD) al ... Full text Cite

Microwave underground propagation and detection

Journal article IEEE Transactions on Microwave Theory and Techniques · March 1, 2002 The detection of buried targets has been a problem of significant interest for decades, with microwave-based sensing constituting an important tool. In this paper, we review the basic issues that characterize microwave-based subsurface sensing. Issues cons ... Full text Cite

Multilevel fast multipole algorithm for general targets on a half-space interface

Journal article IEEE Transactions on Antennas and Propagation · 2002 The multilevel fast multipole algorithm (MLFMA) is considered for scattering from an electrically large conducting or dielectric target resting on the interface of a dielectric half-space. We focus on analysis of the half-space Green's function such that i ... Full text Link to item Cite

Wide-area detection of land mines and unexploded ordnance

Journal article Inverse Problems · 2002 Advanced electromagnetic modelling tools are discussed, focused on sensing surface and buried land mines and unexploded ordnance, situated in a realistic soil environment. The results from these forward models are used to process scattered-field data, for ... Full text Link to item Cite

ICA and PLS modeling for functional analysis and drug sensitivity for DNA microarray signals

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2002 The DNA microarray technique offers an ability to analyze the expression profile of a genome. The complex correlation between the large number of genes present in the genome undermines straightforward understanding of their functionality. In this paper, we ... Full text Cite

Support vector machines for improved multiaspect target recognition using the Fisher kernel scores of Hidden Markov Models

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2002 In conjunction with physics-based feature extraction, Hidden Markov Model (HMM) classifiers have been used successfully to fuse scattering data from multiple target orientations where the target-sensor orientation is generally unknown or "hidden" [1]. The ... Full text Cite

Infrared-image classification using support vector machines

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2002 A target recognition classifier for forward-looking infrared (FLIR) imagery is developed. A target class is defined as a set of contiguous target-sensor orientations (aspects) for which the associated FLIR imagery is stationary. We designed four sets of te ... Full text Cite

Class-based target classification in shallow water channel based on hidden Markov model

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2002 This paper presents a class-based classification approach for targets in a shallow water channel, based on a waveguide propagation model and a Hidden Markov Model (HMM). We utilize the time-frequency properties of wave propagation in a shallow water channe ... Full text Cite

HMM-based multiresolution image segmentation

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2002 A texture segmentation algorithm is developed, utilizing a wavelet-based multi-resolution analysis of general imagery. The wavelet analysis yields a set of quadtrees, each composed of high-high (HH), high-low (HL) and low-high (LH) wavelet coefficients. Hi ... Full text Cite

Parallel extended-born analysis of electromagnetic scattering from 3-Dimensional sub-rough surface targets

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · January 1, 2002 Real 3-dimensional rough surface (RS) half space is treated as a perfect half-space with a special target which leads to the huge problem size if integral equation (IE) based method of moments (MoM) is used, so parallel algorithm is applied to speedup the ... Full text Cite

Efficient time-domain electromagnetic analysis using CDF biorthogonal wavelet expansion

Journal article IEEE MTT S International Microwave Symposium Digest · December 1, 2001 The multi-resolution time-domain (MRTD) algorithm is implemented using Cohen-Daubechies-Feauveau (CDF) wavelet bases, resulting in a computationally efficient numerical scheme for electromagnetic field analysis. The application to a simple scattering probl ... Cite

Multilevel fast multipole algorithm for general dielectric targets in the presence of a lossy half-space

Journal article Radio Science · November 1, 2001 The multilevel fast multipole algorithm (MLFMA) is extended to the problem of an arbitrarily shaped dielectric target in the presence of a lossy, dispersive half-space. The near MLFMA terms are treated rigorously, via a complex-image-technique-based evalua ... Full text Cite

Multiresolution time-domain analysis of plane-wave scattering from general three-dimensional surface and subsurface dielectric targets

Journal article IEEE Transactions on Antennas and Propagation · November 1, 2001 The multiresolution time domain (MRTD) is used to analyze wide-band plane-wave scattering from general dielectric targets embedded in a lossy half-space, with free-space scattering as a special case. A Haar wavelet expansion is used for simplicity, this co ... Full text Cite

Rigorous Modeling of ultrawideband VHF scattering from tree trunks over flat and sloped terrain

Journal article IEEE Transactions on Geoscience and Remote Sensing · October 1, 2001 Three electromagnetic models are employed for the investigation of ultrawideband VHF scattering from tree trunks situated over flat and sloped terrain. Two of the models are numerical, each employing a frequency-domain integral-equation formulation solved ... Full text Cite

Time-domain sensing of targets buried under a Gaussian, exponential, or fractal rough interface

Journal article IEEE Transactions on Geoscience and Remote Sensing · September 1, 2001 We numerically examine subsurface sensing via an ultrawideband ground penetrating radar (GPR) system. The target is assumed to reside under a randomly rough air-ground interface and is illuminated by a pulsed plane wave. The underlying wave physics is addr ... Full text Cite

Genetic algorithm wavelet design for signal classification

Journal article IEEE Transactions on Pattern Analysis and Machine Intelligence · August 1, 2001 Biorthogonal wavelets are applied to parse multiaspect transient scattering data in the context of signal classification. A language-based genetic algorithm is used to design wavelet filters that enhance classification performance. The biorthogonal wavelet ... Full text Cite

Quasi-tem analysis of a novel coplanar waveguide and coupled structure

Journal article Microwave and Optical Technology Letters · July 5, 2001 A novel coplanar waveguide (CPW) and coupled structure with a cylindrical conductor as the shielding are proposed. Simple analytic formulas for the capacitance per unit length and characteristic impedance of the structure are derived using the conformal ma ... Full text Cite

Multiresolution time-domain using CDF biorthogonal wavelets

Journal article IEEE Transactions on Microwave Theory and Techniques · June 18, 2001 A new approach to the multiresolution time-domain (MRTD) algorithm is presented in this paper by introducing a field expansion in terms of biorthogonal scaling and wavelet functions. Particular focus is placed on the Cohen-Daubechies-Feauveau (CDF) biortho ... Full text Cite

Dual hidden Markov model for characterizing wavelet coefficients from multi-aspect scattering data

Journal article Signal Processing · June 1, 2001 Angle-dependent scattering (electromagnetic or acoustic) is considered from a general target, for which the scattered signal is a non-stationary function of the target-sensor orientation. A statistical model is presented for the wavelet coefficients of suc ... Full text Cite

On the wideband EMI response of a rotationally symmetric permeable and conducting target

Journal article IEEE Transactions on Geoscience and Remote Sensing · June 1, 2001 A simple and accurate model is presented for computation of the electromagnetic induction (EMI) resonant frequencies of canonical conducting and ferrous targets, in particular, finite-length cylinders and rings. The imaginary resonant frequencies correspon ... Full text Cite

Multi-aspect detection of surface and shallow-buried unexploded ordnance via ultra-wideband synthetic aperture radar

Journal article IEEE Transactions on Geoscience and Remote Sensing · June 1, 2001 An ultra-wideband (UWB) synthetic aperture radar (SAR) system is investigated for the detection of former bombing ranges, littered by unexploded ordnance (UXO). The objective is detection of a high enough percentage of surface and shallow-buried UXO, with ... Full text Cite

IEEE transactions on geoscience and remote sensing: Foreword

Journal article IEEE Transactions on Geoscience and Remote Sensing · June 1, 2001 Full text Cite

Fast multipole method for scattering from an arbitrary PEC target above or buried in a lossy half space

Journal article IEEE Transactions on Antennas and Propagation · May 1, 2001 The fast multipole method (FMM) was originally developed for perfect electric conductors (PECs) in free space, through exploitation of spectral properties of the free-space Green's function. In the work reported here, the FMM is modified, for scattering fr ... Full text Cite

Multilevel fast multipole algorithm for three-dimensional dielectric targets in the vicinity of a lossy half space

Journal article Microwave and Optical Technology Letters · April 20, 2001 The multilevel fast multipole algorithm (MLFMA) is applied to the problem of a general three-dimensional dielectric target above or below a lossy half space. The dyadic half-space Green's function is evaluated rigorously for the "near" MLFMA interactions, ... Full text Cite

Optimal time-domain detection of a deterministic target buried under a randomly rough interface

Journal article IEEE Transactions on Antennas and Propagation · March 1, 2001 We consider pulsed plane-wave scattering from targets buried under a rough air-ground interface. The properties of the interface are parametrized as a random process with known statistics, and therefore the fields scattered from a particular surface consti ... Full text Cite

An electromagnetic simulation environment

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2001 We present a general purpose simulator that includes electromagnetic scattering tools for buried targets and standard signal processing functionality. Additional modules for genetic or gradient optimization, parallel processing, and multi-aspect target det ... Full text Cite

A comparison of the performance of statistical and fuzzy algorithms for unexploded ordnance detection

Journal article IEEE Transactions on Fuzzy Systems · 2001 In most field environments, unexploded ordnance (UXO) items are found among extensive surface and subsurface clutter and shrapnel from ordnance. Traditional algorithms for UXO remediation experience severe difficulty distinguishing buried targets from anth ... Full text Link to item Cite

Multi-aspect target detection for SAR imagery using hidden Markov models

Journal article IEEE Transactions on Geoscience and Remote Sensing · January 1, 2001 Radar scattering from an illuminated object is often highly dependent on the target-sensor orientation. In typical synthetic aperture radar (SAR) imagery, the information in the multi-aspect target signatures is diffused in the image-formation process. In ... Full text Cite

Efficient evaluation of the half-space Green's function for fast-multipole scattering models

Journal article Microwave and Optical Technology Letters · 2001 There has recently been significant interest in the method-of-moments (MOM) and fast multipole method (FMM) for the analysis of scattering from targets in the presence of a lossy dielectric half space (soil). It is desirable to make the analysis of scatter ... Full text Link to item Cite

A simple preconditioner for electric-field integral equations

Journal article Microwave and Optical Technology Letters · January 1, 2001 A preconditioner is applied to the electric-field integral equation, to improve the convergence of iterative integral-equation solvers, such as the conjugate-gradient (CG) method. The preconditioner accounts for (expansion function)-(testing function) inte ... Full text Cite

Multiaspect classification of airborne targets via physics-based HMMs and matching pursuits

Journal article IEEE Transactions on Aerospace and Electronic Systems · January 1, 2001 Wideband electromagnetic fields scattered from N distinct target-sensor orientations are employed for classification of airborne targets. Each of the scattered waveforms is parsed via physics-based matching pursuits, yielding N feature vectors. The feature ... Full text Cite

Identification of ground targets from sequential HRR radar signatures

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2001 An approach to identifying ground targets from sequential high-range-resolution (HRR) radar signatures is presented. A hidden Markov model (HMM) is employed to model the sequential information contained in multi-aspect target signatures. Dominant range-amp ... Cite

Markov modeling of transient scattering and its application in multi-aspect target classification

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2001 Transient scattered fields from a general target are composed of wavefronts, resonances and time delays, with these constituents linked to the target geometry. A classifier applied transient scattering data requires a statistical model for such fundamental ... Cite

Class-based identification of underwater targets using hidden Markov models

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2001 It has been demonstrated that hidden Markov models (HMMs) provide an effective architecture for classification of distinct targets from multiple target-sensor orientations. In this paper, we present a methodology for designing class-based HMMs that are wel ... Cite

Infrared-image classification using expansion matching filters and hidden Markov trees

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2001 Forward-looking infrared (FLIR) images of targets are characterized by the different target components visible in the image, with such dependent on the target-sensor orientation and target history (i.e., which target components are hot). We define a target ... Cite

Ultrawide-band synthetic aperture radar for detection of unexploded ordnance: Modeling and measurements

Journal article IEEE Transactions on Antennas and Propagation · December 1, 2000 Electromagnetic (EM) scattering from subsurface unexploded ordnance (UXO) is investigated both theoretically and experimentally. Three EM models are considered: the multilevel fast multipole algorithm (MLFMA), the method of moments (MoM), and physical opti ... Full text Cite

Method of moments analysis of a band-pass filter

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · December 1, 2000 In this paper we simulate a band-pass filter, it is a 23 layers multi-layer structure and with 28 vias. The simulation is performed by integration equation based on method of moments (MoM). To speed up the calculation, some new techniques are used. Using t ... Cite

3D surface mesh generator for electromagnetic field problems

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · December 1, 2000 Surface integral equations, which are widely used to solve electromagnetic problems, require mesh generation on the very complex geometry surfaces of the 3D targets and circuits. This is a tedious and time-consuming work, but it is a very important work. A ... Cite

Multi-aspect target classification using hidden Markov models for data fusion

Journal article International Geoscience and Remote Sensing Symposium IGARSS · December 1, 2000 Tactical targets often exhibit a monostatic response that is a function of target-sensor orientation. In SAR image formation, this aspect dependence is lost through integration over the synthetic aperture. The aspect dependent response may be recovered thr ... Cite

Multiresolution time domain analysis of scattering from a rough dielectric surface

Journal article Radio Science · November 1, 2000 The multiresolution time domain (MRTD) algorithm is applied for modeling scattering from a rough dielectric surface. We formulate the Hair MRTD model for an arbitrary interface between two dielectric media. The advantages of this formulation, as compared w ... Full text Cite

Multilevel fast-multipole algorithm for scattering from conducting targets above or embedded in a lossy half space

Journal article IEEE Transactions on Geoscience and Remote Sensing · July 1, 2000 An extension of the multilevel fast multipole algorithm (MLFMA), originally developed for targets in free space, is presented for the electromagnetic scattering from arbitrarily shaped three-dimensional (3-D), electrically large, perfectly conducting targe ... Full text Cite

Hybrid scheme for inverse scattering of electrically large regions

Journal article Radio Science · March 1, 2000 An efficient algorithm is presented for two-dimensional inverse scattering from electrically large regions. The technique is a hybrid combination of the modified-gradient (MG) method and a beam-tracing-based iterative Born method. The beam-tracing-based in ... Full text Cite

Method of moments analysis of electromagnetic scattering from a general three-dimensional dielectric target embedded in a multilayered medium

Journal article Radio Science · March 1, 2000 The method of moments (MOM) is applied to the problem of electromagnetic scattering from general three-dimensional dielectric targets in an arbitrary multilayered environment. The dyadic multilayered Green's function is computed via the method of complex i ... Full text Cite

Multi-aspect target detection for SAR imagery using hidden Markov models and two-dimensional matching pursuits

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 Radar scattering from an illuminated object is often dependent on target-sensor orientation. In synthetic aperture radar (SAR) imagery, the aspect dependence of the target over the aperture is lost during image formation. To recover this directional depend ... Cite

Rigorous electromagnetic modeling of targets concealed in tree foliage

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 A method of moment (MoM) analysis is developed for electromagnetic scattering from a generalized perfectly conducting target in the near field of a tree trunk in a layered medium environment. In this analysis, the tree trunk is modeled as a dielectric body ... Cite

Signal adaptive wavelet design using genetic algorithms

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 While discrete wavelet transforms offer a powerful combination of computational efficiency and compact representation for a broad range of signals, they are often designed without any prior knowledge of the signals under analysis. In this paper, we provide ... Cite

Multi-level fast multipole method for large surface and subsurface targets

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 The multi-level fast multiple algorithm (MLFMA) is applied to the problem of scattering from surface and subsurface targets. In this paper we demonstrate how the MLFMA is modified to handle the half-space problem, and present example results for several sc ... Cite

Time-domain sensing of targets buried under a general rough air-ground interface

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 We numerically examine subsurface sensing via an ultra-wideband ground penetrating radar system. The target is assumed to reside under a randomly rough air-ground interface, and is illuminated by a pulsed plane wave. The underlying wave physics is addresse ... Cite

Dual hidden Markov model characterization of wavelet coefficients from multi-aspect scattering data

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 We consider angle-dependent scattering (electromagnetic or acoustic) from a general target, for which the scattered signal is a non-stationary function of the target-sensor orientation. A statistical model is presented for the wavelet coefficients of such ... Full text Cite

Detection of above ground and subsurface unexploded ordnance using ultra-wideband (UWB) synthetic aperture radar (SAR) and electromagnetic modeling tools

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 2000 Recent development of wideband, high-resolution synthetic aperture radar (SAR) technology has shown that detecting buried targets over large open areas may be possible. Ground clutter and soil type are two limiting factors influencing the practicality of u ... Full text Cite

Calibration of the ARL BoomSAR using rigorous scattering models for fiducial targets over ground

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1999 The Army Research Laboratory (ARL) synthetic aperture radar (SAR) system is placed atop a boom lift, and therefore the system is termed a 'BoomSAR'. The UWB character of this system, covering a frequency spectrum from 40-1200 MHz, makes accurate polarimetr ... Cite

An improved bayesian decision theoretic approach for land mine detection

Journal article IEEE Transactions on Geoscience and Remote Sensing · December 1, 1999 A rigorous signal detection theoretic analysis is used to improve detectability of land mines. The development is performed for sensors that integrate time-domain information to provide a single data point (standard metal detector), those that provide a sa ... Full text Cite

Target identification with wave-based matched pursuits and hidden Markov models

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1999 The method of matched pursuits is an algorithm by which a waveform is parsed into its fundamental constituents here, in the context of short-pulse electromagnetic scattering, wavefronts, and resonances (constituting what we have called wave-based matched p ... Full text Cite

Ultrawideband SAR for detection of subsurface unexploded ordnance (UXO): measurement, modeling and signal processing

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1999 We report here on the use of rigorous scattering models for SAR-based detection of buried UXO. In this paper we concentrate on the algorithm used for the scattering computations, the fast multipole method (FMM), and in the talk we will demonstrate how this ... Cite

Fast multipole method for scattering from an arbitrary perfectly conducting target above or below a lossy half space

Journal article International Geoscience and Remote Sensing Symposium IGARSS · December 1, 1999 The fast multiple method (FMM) was originally developed for perfectly electric conducting (PEC) targets in free space. Here, the FMM is extended to the scattering from a PEC target above or below a lossy half space. The 'near' terms are handled via a metho ... Cite

Optically-activated GaAs switches for ground penetrating radar and firing set applications

Journal article Digest of Technical Papers IEEE International Pulsed Power Conference · December 1, 1999 Optically activated, high gain GaAs switches are being tested for many different applications. Two such applications are ground penetrating radar (GPR) and firing set switches. The ability of high gain GaAs Photoconductive Semiconductor Switches (PCSS) to ... Cite

On the resonances of a dielectric BOR buried in a dispersive layered medium

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1999 A method-of-moments (MoM) analysis is applied to the problem of determining late-time resonances of dielectric bodies of revolution buried in a lossy layered medium, with application to plastic-land-mine identification. To make such an analysis tractable, ... Full text Cite

On the extended-Born technique for scattering from buried dielectric targets

Journal article IEEE Transactions on Antennas and Propagation · November 1, 1999 The extended Born technique is an approximate nonlinear method for analyzing scattering from a weak discontinuity. Moreover, when applied to the low-frequency (electromagnetic induction) applications for which it was developed originally, extended Born has ... Full text Cite

Multiaspect identification of submerged elastic targets via wave-based matching pursuits and hidden Markov models

Journal article Journal of the Acoustical Society of America · August 20, 1999 This paper investigates classification of submerged elastic targets using a sequence of backscattered acoustic signals corresponding to measurements at multiple target-sensor orientations. Wavefront and resonant features are extracted from each of the mult ... Full text Cite

Wide-band electromagnetic scattering from a dielectric BOR buried in a layered lossy dispersive medium

Journal article IEEE Transactions on Antennas and Propagation · April 1999 Full text Cite

Ultra-wide-band synthetic-aperture radar for mine-field detection

Journal article IEEE Antennas and Propagation Magazine · February 1, 1999 A full-wave model is developed for electromagnetic scattering from buried and surface land mines (both conducting and plastic), taking rigorous account of the lossy, dispersive, and potentially layered properties of soil. The (polarimetric) theoretical res ... Full text Cite

Fast multipole method for scattering from 3-D pec targets situated in a half-space environment

Journal article Microwave and Optical Technology Letters · 1999 The fast multipole method (FMM) is extended to the problem of an arbitrary, three-dimensional perfect conductor situated above or below a lossy, dielectric half space. The interactions between basis and testing functions within an FMM cluster, and for near ... Full text Link to item Cite

Multiaspect target identification with wave-based matched pursuits and continuous hidden Markov models

Journal article IEEE Transactions on Pattern Analysis and Machine Intelligence · 1999 Full text Cite

Beam-tracing-based inverse scattering for general aperture antennas

Journal article Journal of the Optical Society of America A Optics and Image Science and Vision · January 1, 1999 Iterative techniques are presented for two-dimensional inverse scattering from electrically large regions. The region is illuminated by transmitters with arbitrary profiles; this is an escalation in complexity from the linesource and the plane-wave excitat ... Full text Cite

Multi-aspect acoustic identification of submerged elastic targets via wave-based matching pursuits and continuous hidden Markov models

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1999 A wave-based matching-pursuits algorithm is used to parse multi-aspect time-domain backscattering data into its underlying wavefront-resonance constituents, or features. Consequently, the N multi-aspect waveforms under test are mapped into N feature vector ... Cite

Short-pulse electromagnetic scattering from arbitrarily oriented subsurface ordnance

Journal article IEEE Transactions on Geoscience and Remote Sensing · January 1, 1999 A rigorous method-of-moments (MoM) analysis is used to model wide-band scattering from general three-dimensional perfectly conducting objects buried in a lossy layered medium. Here, we focus on ordnance buried in a half space (soil). The time-domain fields ... Full text Cite

Wideband frequency- and time-domain EMI for mine detection

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1999 The phenomenology of frequency- and time-domain electromagnetic induction (EMI) is examined in detail, through use of a rigorous electromagnetic scattering model, and through appropriate signal analysis. We demonstrate that both the time- and frequency-dom ... Cite

Hybrid technique combining the moment method with physical optics and uniform asymptotics for scattering from 2-D cylinders

Journal article Microwave and Optical Technology Letters · 1999 Frequency-domain plane-wave scattering from a perfectly conducting two-dimensional cylinder is analyzed by a hybrid formulation combining the method of moments (MoM) and physical optics. Asymptotic techniques are employed to evaluate many of the impedance ... Full text Link to item Cite

On the low-frequency natural response of conducting and permeable targets

Journal article IEEE Transactions on Geoscience and Remote Sensing · 1999 Full text Cite

Performance analysis for radar detection of buried anti-tank and anti-personnel land mines

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1999 A full-wave model is developed for electromagnetic scattering from buried and surface land mines (both conducting and plastic), taking rigorous account of the lossy, dispersive and potentially layered properties of soil. The (polarimetric) theoretical resu ... Cite

Hidden Markov models for multiaspect target classification

Journal article IEEE Transactions on Signal Processing · January 1, 1999 This correspondence presents a new approach for target identification, in which we fuse scattering data from multiple target-sensor orientations. The multiaspect data is processed via hidden Markov model (HMM) classifiers, buttressed by physics-based featu ... Full text Cite

Scattering from complex bodies using a combined direct and iterative technique

Journal article IEEE Transactions on Antennas and Propagation · January 1, 1999 An iterative technique is developed for frequency-domain plane wave scattering from electrically large composite bodies. An electric field integral equation (EFIE) formulation is employed in which the submatrices of the moment-method matrix are uncoupled a ... Full text Cite

Phenomenological modeling for FOPEN SAR: Tree-trunk scattering on flat terrain and with concealed targets

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1999 A method of moments (MoM) analysis is developed for electromagnetic scattering from a dielectric body of revolution (BoR) embedded in a layered medium (the half-space problem constituting a special case). The layered-medium parameters can be lossy and disp ... Cite

Wide-band VHF scattering from a trihedral reflector situated above a lossy dispersive halfspace

Journal article IEEE Transactions on Geoscience and Remote Sensing · 1999 Full text Cite

Multi-aspect target detection in SAR imagery using hidden Markov models

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1999 It is well known that radar scattering from an illuminated object is often dependent on target-sensor orientation. In typical synthetic aperture radar (SAR) imagery, such aspect dependence is lost during image formation. We apply a sequence of directional ... Cite

Physics-based classification of targets in SAR imagery using subaperture sequences

Journal article ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 1999 It is well known that radar scattering from an illuminated object is often highly aspect dependent. We have developed a multi-aspect target classification technique for SAR imagery that incorporates matching-pursuits feature extraction from each of a seque ... Full text Cite

Ultrawideband scattering from and the resonances of buried dielectric mines

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1998 A method of moments (MoM) analysis has been developed for the calculation of electromagnetic scattering from and the natural resonances of a dielectric body of revolution (BOR) embedded in a layered medium (the half-space problem constituting a special cas ... Full text Cite

Analysis and processing of ultra wide-band SAR imagery for buried landmine detection

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1998 Experimental and theoretical results are presented for ultra wide-band (UWB) synthetic aperture radar (SAR) signatures of buried anti-tank and anti-personnel mines. Such are characterized by resonancelike peaks as well as valleys, across the 50-1200 MHz ba ... Full text Cite

Comparison of model-based results with measured data for metal buried mines

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1998 To detect and identify buried mines, the U.S. Army Research Laboratory (ARL) is using its ultra wideband (UWB) radar in a ground-penetrating mode. Operating in the frequency band from 50 to 1200 MHz, the radar is mounted on a mobile boom lift platform (Boo ... Full text Cite

Mode coupling and leakage effects in finite-size printed interconnects

Journal article IEEE Transactions on Microwave Theory and Techniques · December 1, 1998 A multimode analysis is used to describe how leakage effects are manifested in general printed interconnects situated on substrates of finite size. In the vicinity of discrete frequencies, it is shown that the analysis reduces to classical coupled-mode the ... Full text Cite

Multiresolution signature-based sar target detection

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1998 A full-wave electromagnetic-scattering model is utilized to effect a land-mine detector via a multiresolution template-matching-like algorithm. Detection is performed on fully polarimetric ultra-wideband (50-1200 MHz) synthetic aperture radar (SAR) imagery ... Full text Cite

A hybrid (parabolic equation)-(gaussian beam) algorithm for wave propagation through large inhomogeneous regions

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1998 The wide-angle split-step parabolic equation (PE) algorithm is used to model electromagnetic wave propagation over general inhomogeneous terrain up to a height h. The PE-computed fields at h are then projected onto a complete Gabor basis from which we effe ... Full text Cite

Polarimetric SAR imaging of buried landmines

Journal article IEEE Transactions on Geoscience and Remote Sensing · December 1, 1998 If the fields incident on a buried body of revolution are polarized vertically or horizontally (relative to the ground), the backscattered fields are exclusively copolarized (i.e., there are no cross-polarized backscattered fields). After substantiating th ... Full text Cite

Subbanding of temporal and spatial UWB SAR imagery of buried mines

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1998 A numerical algorithm has been developed for the modeling of ultra-wideband (UWB) plane-wave scattering from a class of buried mines. In particular, the model assumes that a mine can be simulated as a body of revolution (BOR). The numerical results indicat ... Full text Cite

Time-domain sensing of targets buried under a rough air-ground interface

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1998 We consider plane wave time-domain scattering from a fixed target in the presence of a rough (random) surface with application to ground penetrating radar. The timedomain scattering data are computed via a two-dimensional (2-D) finite-difference time-domai ... Full text Cite

Ultra-wideband synthetic aperture radar for mine field detection

Journal article Ultra Wideband Short Pulse Electromagnetics 4 · January 1, 1998 Ground-penetrating radar (GPR) constitutes one of the oldest technologies for subsurface sensing. Most of such systems are placed in direct or near-direct contact with the earth surface. A significant drawback of this approach is the lack of «standoff», a ... Full text Cite

Wave-based matching-pursuits detection of submerged elastic targets

Journal article Journal of the Acoustical Society of America · January 1, 1998 Matching pursuits is a nonlinear algorithm which iteratively projects a given signal onto a complete dictionary of vectors. The dictionary is constructed such that it is well matched to the signals of interest and poorly matched to the noise, thereby affor ... Full text Cite

Short-pulse scattering from a dielectric BOR buried in a lossy layered medium

Journal article International Geoscience and Remote Sensing Symposium IGARSS · January 1, 1998 A method of moments (MoM) analysis has been developed for ultra-wideband (UWB) electromagnetic scattering from a dielectric body of revolution (BOR) embedded in a layered medium. The layered-medium can be lossy and dispersive, of interest for simulating so ... Full text Cite

Resonances of a dielectric BOR buried in a lossy, dispersive layered medium

Journal article International Geoscience and Remote Sensing Symposium IGARSS · January 1, 1998 A method-of-moments (MoM) analysis is used to determine late-time resonances of dielectric bodies of revolution (BOR), buffed in a lossy layered medium, with application to plastic-land-mine identification. To make the analysis tractable, we have extended ... Full text Cite

Ultra-wideband, short-pulse ground-penetrating radar: simulation and measurement

Journal article IEEE Transactions on Geoscience and Remote Sensing · December 1, 1997 Ultra-wideband (UWB), short-pulse (SP) radar is investigated theoretically and experimentally for the detection and identification of targets buried in and placed atop soil. The calculations are performed using a rigorous, three-dimensional (3-D) Method of ... Full text Cite

Wave-oriented signal processing of dispersive time-domain scattering data

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1997 Phase-space data processing is receiving increased attention because of its potential for furnishing new discriminants relating to classification and identification of targets and other scattering environments. Primary emphasis has been on time-frequency p ... Full text Cite

On the superresolution identification of observables from swept-frequency scattering data

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1997 A superresolution signal processing algorithm is used for the identification of wavefronts from the fields scattered from several canonical targets. Particular wave objects that are examined are single and multiple edge diffraction, scattering from flat an ... Full text Cite

Matching pursuits with a wavebased dictionary

Journal article IEEE Transactions on Signal Processing · December 1, 1997 The method of matching pursuits utilizes a nonlinear iterative procedure to project a given waveform onto a particular dictionary. For scattering problems, the most appropriate dictionary is composed of waveobjects that are consistent with the underlying w ... Cite

Time-domain characteristics of slotted-waveguide leaky-wave antennas

Journal article IEEE Microwave and Guided Wave Letters · May 1, 1997 A general asymptotic theory is presented to describe the time-domain behavior of leaky transmission lines and antennas. The results are interpreted via a simple geometric construct and data are presented for the particular case of time-domain radiation fro ... Full text Cite

Matching pursuits with a wave-based dictionary

Journal article IEEE Transactions on Signal Processing · January 1, 1997 The method of matching pursuits utilizes a nonlinear iterative procedure to project a given waveform onto a particular dictionary. For scattering problems, the most appropriate dictionary is composed of waveobjects that are consistent with the underlying w ... Full text Cite

Resonances of perfectly conducting wires and bodies of revolution buried in a lossy dispersive half-space

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1996 The method of moments (MoM) is utilized to compute the complex resonant frequencies and modal currents of perfectly conducting wires and bodies of revolution buried in a lossy dispersive half space. To make such an analysis tractable computationally, the h ... Full text Cite

Short-pulse plane-wave scattering from buried perfectly conducting bodies of revolution

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1996 The method of moments is used to analyze shortpulse plane-wave scattering from perfectly conducting bodies of revolution buried in a lossy, dispersive half space. The analysis is performed in the frequency domain, with the time-domain fields synthesized vi ... Full text Cite

Picosecond-pulse and millimeter-wave spectroscopy of barium ferrite

Journal article IEEE Transactions on Magnetics · December 1, 1996 Transmission measurements of a barium ferrite pressed-powder sample have been made with an optically switched picosecond-pulse spectrometer. By comparison to millimeter-wave spectroscopy of the same sample, features related to the ferrimagnetic resonance h ... Full text Cite

High-frequency fields excited by truncated arrays of nonuniformly distributed filamentary scatterers on an infinite dielectric slab: parameterizing (leaky mode)-(floquet mode) interaction

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1996 In previous studies, we have developed and tested observable-based parameterization (OBP) of time-harmonic wavefield scattering by periodic or aperiodic finite arrays of planar strip and filament scatterers. The resulting algorithm is based on truncated Fl ... Full text Cite

Wave-oriented processing of scattered field data from a plane-wave-excited finite array of filaments on an infinite dielectric slab

Journal article IEEE Transactions on Antennas and Propagation · December 1, 1996 In a companion paper, we presented the formulation and solution for time-harmonic plane wave fields scattered by truncated periodic and aperiodic arrays of infinitely long filaments on an infinite dielectric slab. The solution was constructed so as to high ... Full text Cite

Resonances of buried targets [radar theory]

Journal article Proc. SPIE - Int. Soc. Opt. Eng. (USA) · 1996 The method of moments is utilized to compute the complex resonant frequencies and modal currents of perfectly conducting wires and bodies of revolution buried in a lossy, dispersive half space. To make such an analysis tractable computationally, the half-s ... Cite

Short-pulse scattering from buried wires and bodies of revolution [ground penetrating radar]

Journal article Proc. SPIE - Int. Soc. Opt. Eng. (USA) · 1996 The method of moments is used to analyze short-pulse plane-wave scattering from perfectly conducting thin wires and bodies of revolution buried in a lossy, dispersive half space. The analysis is performed in the frequency domain, with the time-domain field ... Cite

FDTD analysis of plane-wave diffraction from microwave devices on an infinite dielectric slab

Journal article IEEE Microwave and Guided Wave Letters · January 1, 1996 A 2-D (two-dimensional) Huygens surface is developed for the finite difference time domain (FDTD) algorithm, allowing the investigation of pulsed plane-wave scattering from arbitrary 2-D structures placed on or in an infinite dielectric slab. Example resul ... Full text Cite

Resonances of buried targets

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1996 The Method of Moments is utilized to compute the complex resonant frequencies and modal currents of perfectly conducting wires and bodies of revolution buried in a lossy, dispersive half space. To make such an analysis tractable computationally, the half-s ... Cite

Short-pulse scattering from buried wires and bodies of revolution

Journal article Proceedings of SPIE the International Society for Optical Engineering · January 1, 1996 The method of moments is used to analyze short-pulse plane- wave scattering from perfectly conducting thin wires and bodies of revolution buried in a lossy, dispersive half space. The analysis is performed in the frequency domain, with the time-domain fiel ... Cite

Dispersion curves and transmission spectra of a two-dimensional photonic band-gap crystal: Theory and experiment

Journal article Journal of Applied Physics · December 1, 1995 An on-shell method that combines plane-wave and finite-difference techniques for the calculation of dispersion curves and transmission spectra for electromagnetic fields in photonic band-gap crystals is presented. The overall problem is decomposed into a f ... Full text Cite

Model-based object recognition by wave-oriented data processing

Journal article Proceedings of SPIE the International Society for Optical Engineering · December 1, 1995 Object recognition can be parametrized systematically through physically robust wave objects by linking features (observables) in scattered field data with features on the object (target) giving rise to the data. The wave objects are broadly separated into ... Cite

Moment-method modeling of short-pulse scattering from and the resonances of a wire buried inside a lossy, dispersive half-space

Journal article IEEE Transactions on Antennas and Propagation · January 1, 1995 A frequency-domain method-of-moments (MoM) algorithm is used to model short-pulse plane-wave scattering from a wire buried inside a lossy, dispersive half-space with the time-domain scattered fields computed via Fourier transform. Further, the complex reso ... Full text Cite

Mode Conversion and Leaky-Wave Excitation at Open-End Coupled-Microstrip Discontinuities

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1995 The method of moments (MoM) is used to study mode conversion and leaky-wave excitation at an asymmetric coupledmicrostrip discontinuity. The results show that significant mode conversion can occur at such discontinuities and that dominant leaky-wave modes ... Full text Cite

Short-Pulse Propagation in a Hollow Waveguide: Analysis, Optoelectronic Measurement, and Signal Processing

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1995 An asymptotic analysis is performed for short-pulse propagation in a hollow waveguide. It is demonstrated that each time-domain mode supported by the guide is characterized by a time-dependent frequency which, as time proceeds, approaches the modal cutoff ... Full text Cite

Mode conversion and leaky-wave excitation at open-end coupled microstrip discontinuities

Journal article IEEE MTT S International Microwave Symposium Digest · January 1, 1995 The method of moments (MoM) is used to study mode conversion and leaky-wave excitation at an asymmetric coupled-microstrip discontinuity. The results show that significant mode conversion can occur at such discontinuities and that fundamental leaky-wave mo ... Cite

Photoconductively switched antennas for measuring target resonances

Journal article Applied Physics Letters · December 1, 1994 Coplanar-strip horn antennas are switched photoconductively to generate picosecond bursts of freely propagating electromagnetic energy with bandwidth covering 15-75 GHz. The antennas are fabricated on GaAs grown by molecular beam epitaxy at low substrate t ... Full text Cite

Frequency domain wave-oriented data processing of scattering by truncated periodic strip gratings

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · December 1, 1994 Guided by a Floquet-modified GTD model developed by us recently for finite periodic and weakly aperiodic wire gratings, we propose and apply space-wavenumber phase space processing algorithms to extract that phenomenology from a numerical data base. A Meth ... Cite

High frequency radiation from truncated arrays of filamentary sources on a dielectric slab

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · December 1, 1994 Previously explored radiation phenomenologies pertaining to individual line sources on a dielectric slab and to finite arrays of line sources in free space are here combined to synthesize and parametrize radiation from finite line source arrays located on ... Cite

Time domain wave-oriented data processing of scattering by truncated periodic strip gratings

Journal article IEEE Antennas and Propagation Society AP S International Symposium Digest · December 1, 1994 In a companion paper at this meeting, we have investigated wave-oriented processing techniques which extract from frequency domain (FD) scattering data for truncated periodic strip gratings the wave phenomenology that ties features in data to scattering me ... Cite

Wave-Oriented Data Processing for Frequency and Time-Domain Scattering by Nonuniform Truncated Arrays

Journal article IEEE Antennas and Propagation Magazine · January 1, 1994 Full text Cite

Scattering by 2d strips using an asymptotic hybrid formulation combining the method‐of‐moments and physical‐optics techniques

Journal article Microwave and Optical Technology Letters · January 1, 1994 Frequency‐domain plane‐wave scattering from perfectly conducting two‐dimensional strips is analyzed by a hybrid formulation combining the method of moments (MOM) and physical optics (PO). Asymptotic techniques are employed to evaluate many of the impedance ... Full text Cite

Diffraction theory of frequency- and time-domain scattering by weakly aperiodic truncated thin-wire gratings

Journal article Journal of the Optical Society of America A Optics and Image Science and Vision · January 1, 1994 An arbitrarily illuminated truncated nonuniform thin-wire grating produces a scattered field that can be synthesized by superposition of the fields radiated by the currents induced on each wire element. For weak departures from periodicity and for quasi-pl ... Full text Cite

Wideband Dispersion Measurements of Water in Reflection and Transmission

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1994 Planar antennas are switched photoconductively to generate picosecond bursts of freely-propagating radiation with usable spectral amplitudes from 5 to 85 GHz. This radiation is used to perform reflection and transmission measurements on materials, with exp ... Full text Cite

Frequency and time domain Bragg-modulated ray acoustics for truncated periodic arrays

Journal article Journal of the Acoustical Society of America · January 1, 1994 Many scenarios in underwater acoustics involve radiation from, or scattering by, configurations with periodic or quasiperiodic features. Depending on the operating conditions, the acoustic field generated by these processes carries the gross imprint of per ... Full text Cite

Time-domain wave-oriented data processing of scattering by nonuniform truncated gratings

Journal article Journal of the Optical Society of America A Optics and Image Science and Vision · January 1, 1994 In a companion paper [J. Opt. Soc. Am. A 11, 2675 (1994)] we investigated wave-oriented processing techniques that extract from frequency-domain (FD) scattering data for nonuniform truncated thin-wire or strip gratings the wave phenomenology that ties feat ... Full text Cite

Frequency-domain scattering by nonuniform truncated arrays: Wave-oriented data processing for inversion and imaging

Journal article Journal of the Optical Society of America A Optics and Image Science and Vision · January 1, 1994 We previously presented an asymptotic diffraction theory for time-harmonic and transient scattering by arbitrarily illuminated truncated nonuniform thin-wire gratings [J. Opt. Soc. Am. A 11, 1291 (1994)]. We parameterized and interpreted the results in ter ... Full text Cite

Dispersive Modes in the Time Domain: Analysis and Time-Frequency Representation

Journal article IEEE Microwave and Guided Wave Letters · January 1, 1994 Four algorithms for time-frequency (TF) distributions are considered for the processing and interpretation of dispersive time-domain (TD) data: The short-time Fourier transform, frequency and time-domain wavelets, and a new ARMA-based representation. The T ... Full text Cite

Short-pulse scattering measurements from dielectric spheres using photoconductively switched antennas

Journal article Applied Physics Letters · December 1, 1993 Planar antennas are switched photoconductively using optical pulses generated by a picosecond laser system. The freely propagating radiation, of picosecond duration and with bandwidth extending from 5 to 75 GHz, is used to perform short-pulse scattering me ... Full text Cite

Polymer composites

Journal article Polymer News · 1993 Cite

Wave-orientated processing of scattering data

Journal article Electronics Letters · January 1, 1993 Windowed transforms applied to scattering data gathered along an elevated track parallel to a scattering surface are shown to provide local plane wave spectra which can be backpropagated to synthesise distinct features of the scattering environment. The me ... Full text Cite

Efficient analytical‐numerical modelling of ultra‐wideband pulsed plane wave scattering

Journal article International Journal of Numerical Modelling Electronic Networks Devices and Fields · January 1, 1993 Ultra‐wideband (UWB) pulsed plane wave scattering from a large but finite strip grating in free space is analysed in the frequency domain via decomposition into plane wave spectra, implemented numerically by the method of moments, and then inverted into th ... Full text Cite

Time Harmonic and Transient Scattering by Finite Periodic Flat Strip Arrays: Hybrid (Ray)-(Floquet Mode)-(MOM) Algorithm

Journal article IEEE Transactions on Antennas and Propagation · January 1, 1993 Finite periodic structures are of interest in a variety of Narrowband applications. With the trend toward wider bandwidth, culiminating in the ultra-wideband or short pulse (SP) regime, it is of interest to explore how well defined narrowband wave fields, ... Full text Cite

Characterization of Layered Dielectrics with Short Electromagnetic Pulses

Journal article IEEE Journal of Quantum Electronics · January 1, 1993 Picosecond duration bursts of electromagnetic radiation are used to study short-pulse wave propagation in layered dielectric materials. The radiation is generated using planar antennas that are switched photoconductively. The measurements are compared with ... Full text Cite

Characterization of Planar Antennas Fabricated on GaAs Epilayers Containing As Clusters for Picosecond Short-Pulse Applications

Journal article IEEE Microwave and Guided Wave Letters · January 1, 1993 Coplanar-strip horn antennas are fabricated on GaAs grown by molecular beam epitaxy at substrate temperatures of 220, 250, and 270° C. These antennas are switched photo conductively using a picosecond laser to generate and detect freely propagating bursts ... Full text Cite

Ultra-wideband transient microwave scattering measurements using optoelectronically switched antennas

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1993 Ultra-wideband transient microwave scattering measurements are performed using optoelectronically switched planar antennas. The laser-based system produces freely propagating bursts of picosecond duration electromagnetic radiation, with a bandwidth extendi ... Full text Cite

Time-Domain Design-Oriented Parametrization of Truncated Periodic Strip Gratings

Journal article IEEE Microwave and Guided Wave Letters · January 1, 1993 Asymptotic methods are used to develop an algorithm that parametrizes time-domain plane-wave interaction with a truncated grating of periodically spaced, perfectly conducting strips in free space. By distinctly displaying the edge effects as well as the tr ... Full text Cite

Ultra-wideband characterization of lossy materials: short-pulse microwave measurements

Journal article IEEE MTT S International Microwave Symposium Digest · January 1, 1993 Planar antennas are switched photoconductively to generate picosecond bursts of freely-propagating radiation with usable spectral amplitudes in the 5 to 85 GHz frequency range. This radiation is used to measure the frequency-dependent, complex index of ref ... Cite

Packaged printed transmission lines: modal phenomena and relation to leakage

Journal article IEEE MTT S International Microwave Symposium Digest · January 1, 1993 The dispersion curves of the modes on shielded printed transmission lines often interact with the dispersion curves of box (package) guided modes in a classical coupled-mode manner. It is shown here that this effect is related directly to the phenomenon of ... Cite

Efficient Computation of High-Frequency Two-Dimensional Effects in Multiconductor Printed Interconnects

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1992 The spectral domain technique with a Galerkin moment method solution is used to study high-frequency, two-dimensional effects such as dispersion and leakage in multiconductor printed interconnects. A simple asymptotic procedure is used to significantly imp ... Full text Cite

Leaky Waves on Broadside-Coupled Microstrip

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1992 Broadside-coupled microstrip with and without conducting side walls are studied using a full-wave spectral-domain analysis. Special attention is directed towards possible leakage to the parallel plate mode and its potential effects in practical integrated ... Full text Cite

Design-Oriented Parametrization of Truncated Periodic-Strip Gratings

Journal article IEEE Microwave and Guided Wave Letters · January 1, 1992 Spectral domain asymptotics are used to develop a hybrid (ray)-(Floquet mode) parametrization that models time-harmonic plane-wave interaction with a truncated grating of periodically spaced, coplanar, infinitesimally thin, perfectly conducting strips in f ... Full text Cite

Leakage effects in broadside-coupled microstrip

Journal article IEEE MTT S International Microwave Symposium Digest · December 1, 1991 Broadside-coupled microstrip with and without conducting side walls is studied using a full-wave spectral-domain analysis. Special attention is directed towards possible leakage to the parallel plate mode and its potential effects in practical integrated c ... Cite

A Study of Slotline Leaky-Wave Antennas

Journal article IEEE Transactions on Antennas and Propagation · January 1, 1990 The results of an experimental investigation of several millimeter-wave slotline leaky-wave antennas at 94 GHz are given. The applicability of a simple model for the slot field is discussed together with design criteria for such antennas. © 1990 IEEE ... Full text Cite

Pulse Propagation on Multi-Layered Circuit-Level Interconnects

Journal article Journal of Electromagnetic Waves and Applications · January 1, 1990 High-speed pulse propagation on terminated dual-level interconnects is investigated. The interconnects, modeled as multiconductor microstrip embedded in a layered dielectric substrate, are analyzed by using a full-wave approach to calculate dispersive eige ... Full text Cite

Isolation Effects in Single- and Dual-Plane VLSI Interconnects

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1990 The issue of interline coupling in high-speed VLSI interconnects is addressed. A full-wave-based technique is used to numerically solve for the modes and hence the line voltages and currents for multiconductor microstrip. The accuracy of these results is c ... Full text Cite

Modal transition phenomena in shielded microstrip with anisotropic substrates

Journal article IEEE MTT S International Microwave Symposium Digest · January 1, 1990 Modal transitions involving the quasi-TEM (transverse electromagnetic) mode and higher-order modes in shielded microstrip and suspended microstrip with anisotropic substrates are studied. For the class of anisotropy studied, the largest eigenvalue may not ... Cite

Characteristic impedance of multilevel, multiconductor hybrid mode microstrip

Journal article IEEE Transactions on Magnetics · January 1, 1989 Two definitions of modal characteristic impedance for multiconductor, hybrid-mode microstrip are compared. The sensitivity of each to numerical inaccuracies is discussed. The plausibility of negative values of modal characteristic impedance is shown for tw ... Full text Cite

An Equivalent Circuit Model for Terminated Hybrid-Mode Multiconductor Transmission Lines

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1989 An equivalent circuit for terminated hybrid-mode multiconductor transmission lines is presented. Existing CAD packages, such as SPICE, can be used for its implementation. Model parameters can be found from either a TEM or a full-wave analysis of the transm ... Full text Cite

Matched Windows in Circular Waveguide

Journal article IEEE Transactions on Microwave Theory and Techniques · January 1, 1988 Design curves are presented for the matching of a dielectric window in circular waveguide propagating the dominant TE11 mode. The matching is accomplished by thick or thin inductive irises which are in contact with the window on both sides. This configurat ... Full text Cite

NUMERICAL ANALYSIS OF PLANAR HIGH FREQUENCY INTEGRATED CIRCUIT GEOMETRIES.

Journal article undefined · December 1, 1987 Formulation procedures together with numerical results are presented for a variety of planar transmission line geometries frequently encountered in high-frequency integrated circuits. Data for a full-wave example are given for a coupled-line microstrip int ... Cite

ANALYSIS OF VLSI INTERCONNECT STRUCTURES.

Journal article IEEE MTT S International Microwave Symposium Digest · January 1, 1987 A typical VLSI circuit contains a large number of devices with planar metallic interconnects between them. Frequently, these interconnects are in the form of microstrip, analogous to what might be used in a millimeter-wave or microwave integrated circuit. ... Full text Cite