Kaizhu Huang

Journal article Pattern Recognition · October 1, 2026 With the advancement of large language models (LLMs) and multimodal LLMs (MLLMs), table reasoning has achieved significant progress. However, most existing works focus predominantly on textual or rendered tables, which differ substantially from real-world ... Full text Cite

Diff-Oracle: Learning Styles and Contents to Augment Realistic Oracle Characters in Diffusion Model

Journal article ACM Transactions on Multimedia Computing Communications and Applications · June 21, 2026 Recognizing oracle bone scripts plays an important role in Chinese archaeology and philology. However, a significant challenge remains because of the scarcity of oracle character images. To overcome this issue, we propose Diff-Oracle, a novel multi-modal c ... Full text Cite

Switch, Reason, and Revise: Enhancing Reasoning Capability of Video Game AI by Large Language Models

Journal article IEEE Transactions on Games · June 1, 2026 Attributing to the strong reasoning capability, behavior models in artificial intelligence (AI) for games play a crucial role in creating gaming experiences. For further enhancing the reasoning capability of behavior models, we propose a novel switch, reas ... Full text Cite

You look from old classes: Towards accurate few shot class-incremental learning

Journal article Pattern Recognition · April 1, 2026 Few-shot class incremental learning (FSCIL) is a common but difficult task that faces two challenges: catastrophic forgetting of old classes and insufficient learning of new classes with limited samples. Recent wisdom focuses on preventing catastrophic for ... Full text Cite

TPGCA: Transferable Policy Generation and Credit Assignment Network for Cooperative Multiagent Reinforcement Learning

Journal article IEEE Transactions on Computational Social Systems · April 1, 2026 Multiagent reinforcement learning (MARL) methods have good application performances and prospects in cooperative tasks. To improve the capability of agent policy learning in new scenarios, some methods transfer the learned policy knowledge to new scenarios ... Full text Cite

IDEA: Image description enhanced CLIP-adapter for image classification

Journal article Pattern Recognition · March 1, 2026 CLIP (Contrastive Language-Image Pre-training) has attained great success in pattern recognition and computer vision. Transferring CLIP to downstream tasks (e.g., zero- or few-shot classification) is a hot topic in multimodal learning. However, current stu ... Full text Cite

MedMAP: Promoting Incomplete Multi-Modal Brain Tumor Segmentation With Alignment.

Journal article IEEE journal of biomedical and health informatics · March 2026 Brain tumor segmentation is often based on multiple magnetic resonance imaging (MRI). However, in clinical practice, certain modalities of MRI may be missing, which presents a more difficult scenario. To cope with this challenge, Knowledge Distillation, Do ... Full text Cite

Rethinking Spectral Graph Neural Networks With Spatially Adaptive Filtering.

Journal article IEEE transactions on neural networks and learning systems · March 2026 Whilst spectral graph neural networks (GNNs) are theoretically well-founded in the spectral domain, their practical reliance on polynomial approximation implies a profound linkage to the spatial domain. As previous studies rarely examine spectral GNNs from ... Full text Cite

Point2pix-Zero: Point-driven refined diffusion for multi-object image editing

Journal article Pattern Recognition · February 1, 2026 Semantic image editing methods employing large-scale diffusion models have made significant strides in precise and controlled image editing with text prompts as guidance. However, these models struggle to handle complex images containing hard-described obj ... Full text Cite

Lena-TRNN: Exploring energy flow for time series prediction.

Journal article Neural networks : the official journal of the International Neural Network Society · February 2026 We focus on exploring the inherent energy flow for time series prediction in this paper, i.e., we consider the inherent energy of time series data as a sequence measuring properties such as fluctuations, oscillations, and trends. Distinctive with main-stre ... Full text Cite

Learning transferable collaborative behaviors for multiple agents in the game environment

Journal article Entertainment Computing · January 1, 2026 Cooperative Multi-Agent Reinforcement Learning (CMARL) enables multiple agents to learn collaborative policies for accomplishing complex tasks in virtual game environments. However, most CMARL algorithms have difficulties in learning effective collaborativ ... Full text Cite

Consensus-Based Communication for Cooperative Multi-Agent Reinforcement Learnin

Journal article IEEE Transactions on Cognitive and Developmental Systems · 2026 Full text Cite

Prompt-Enhanced: Leveraging language representation for prompt continual learning.

Journal article Neural networks : the official journal of the International Neural Network Society · October 2025 Continual learning enables models to learn from an evolving stream of data without forgetting the previously acquired skills. Traditional methods often rely on the rehearsal buffers or extended network structures to retain the previous knowledge, which inc ... Full text Cite

SCMix: Stochastic Compound Mixing for Open Compound Domain Adaptation in Semantic Segmentation.

Journal article IEEE transactions on neural networks and learning systems · September 2025 Open compound domain adaptation (OCDA) aims to transfer knowledge from a labeled source domain to a mix of unlabeled homogeneous compound target domains while generalizing to open unseen domains. Existing OCDA methods solve the intradomain gaps by a divide ... Full text Cite

Stagger Network: Rethinking information loss in medical image segmentation with various-sized targets.

Journal article Neural networks : the official journal of the International Neural Network Society · August 2025 Medical image segmentation presents the challenge of segmenting various-size targets, demanding the model to effectively capture both local and global information. Despite recent efforts using CNNs and ViTs to predict annotations of different scales, these ... Full text Cite

Revisiting 3D point cloud analysis with Markov process

Journal article Pattern Recognition · February 1, 2025 3D point cloud analysis has recently garnered significant attention due to its capacity to provide more comprehensive information compared to 2D images. To confront the inherent irregular and unstructured properties of point clouds, recent research efforts ... Full text Cite

HDMTK: Full Integration of Hierarchical Decision-Making and Tactical Knowledge in Multiagent Adversarial Games

Journal article IEEE Transactions on Cognitive and Developmental Systems · January 1, 2025 In the field of adversarial games, existing decision-making algorithms primarily rely on reinforcement learning, which can theoretically adapt to diverse scenarios through trial and error. However, these algorithms often face the challenges of low effectiv ... Full text Cite

Distillation-Based Domain Generalization for Cross-Dataset EEG-Based Emotion Recognition

Journal article IEEE Transactions on Emerging Topics in Computational Intelligence · January 1, 2025 Electroencephalogram (EEG)-based emotion recognition has gradually become a research hotspot with extensive real-world applications. Differences in EEG signals across subjects usually lead to the unsatisfactory performance in subject-independent emotion re ... Full text Cite

Open-Pose 3D zero-shot learning: Benchmark and challenges.

Journal article Neural networks : the official journal of the International Neural Network Society · January 2025 With the explosive 3D data growth, the urgency of utilizing zero-shot learning to facilitate data labeling becomes evident. Recently, methods transferring language or language-image pre-training models like Contrastive Language-Image Pre-training (CLIP) to ... Full text Cite

ES-GNN: Generalizing Graph Neural Networks Beyond Homophily With Edge Splitting.

Journal article IEEE transactions on pattern analysis and machine intelligence · December 2024 While Graph Neural Networks (GNNs) have achieved enormous success in multiple graph analytical tasks, modern variants mostly rely on the strong inductive bias of homophily. However, real-world networks typically exhibit both homophilic and heterophilic lin ... Full text Cite

Biomedical Information Retrieval with Positive-Unlabeled Learning and Knowledge Graphs

Journal article ACM Transactions on Intelligent Systems and Technology · November 4, 2024 The rapid growth of biomedical publications has presented significant challenges in the field of information retrieval. Most existing work focuses on document retrieval given explicit queries. However, in real applications such as curated biomedica ... Full text Cite

A generalizable framework for low-rank tensor completion with numerical priors

Journal article Pattern Recognition · November 1, 2024 Low-Rank Tensor Completion, a method which exploits the inherent structure of tensors, has been studied extensively as an effective approach to tensor completion. Whilst such methods attained great success, none have systematically considered exploiting th ... Full text Cite

EgPDE-Net: Building Continuous Neural Networks for Time Series Prediction With Exogenous Variables.

Journal article IEEE transactions on cybernetics · September 2024 While exogenous variables have a major impact on performance improvement in time series analysis, interseries correlation and time dependence among them are rarely considered in the present continuous methods. The dynamical systems of multivariate time ser ... Full text Cite

SaliencyCut: Augmenting plausible anomalies for anomaly detection

Journal article Pattern Recognition · September 1, 2024 Anomaly detection under the open-set scenario is a challenging task that requires learning discriminative features to detect anomalies that were even unseen during training. As a cheap yet effective approach, data augmentation has been widely used to creat ... Full text Cite

Absorb and Repel: Pseudo-Label Refinement for Intra-Camera Supervised Person Re-Identification

Journal article IEEE Transactions on Artificial Intelligence · June 1, 2024 Person re-identification (ReID) aims to identify pedestrian images with the same identity across non-overlapping camera views. Intra-camera supervised person re-identification (ICS-ReID) is a new paradigm that trains a model using only intra-camera labels, ... Full text Cite

Continuous Image Outpainting with Neural ODE

Journal article ACM Transactions on Multimedia Computing Communications and Applications · April 25, 2024 Generalised image outpainting is an important and active research topic in computer vision, which aims to extend appealing content all-side around a given image. Existing state-of-the-art outpainting methods often rely on discrete extrapolation to extend t ... Full text Cite

Perturbation diversity certificates robust generalization.

Journal article Neural networks : the official journal of the International Neural Network Society · April 2024 Whilst adversarial training has been proven to be one most effective defending method against adversarial attacks for deep neural networks, it suffers from over-fitting on training adversarial data and thus may not guarantee the robust generalization. This ... Full text Cite

Instance-Specific Model Perturbation Improves Generalized Zero-Shot Learning.

Journal article Neural computation · April 2024 Zero-shot learning (ZSL) refers to the design of predictive functions on new classes (unseen classes) of data that have never been seen during training. In a more practical scenario, generalized zero-shot learning (GZSL) requires predicting both seen and u ... Full text Cite

Can Perturbations Help Reduce Investment Risks? Risk-aware Stock Recommendation via Split Variational Adversarial Training

Journal article ACM Transactions on Information Systems · March 22, 2024 In the stock market, a successful investment requires a good balance between profits and risks. Based on the learning to rank paradigm, stock recommendation has been widely studied in quantitative finance to recommend stocks with higher return ratios for i ... Full text Cite

Learning Disentangled Graph Convolutional Networks Locally and Globally.

Journal article IEEE transactions on neural networks and learning systems · March 2024 Graph convolutional networks (GCNs) emerge as the most successful learning models for graph-structured data. Despite their success, existing GCNs usually ignore the entangled latent factors typically arising in real-world graphs, which results in nonexplai ... Full text Cite

PointNu-Net: Keypoint-Assisted Convolutional Neural Network for Simultaneous Multi-Tissue Histology Nuclei Segmentation and Classification

Journal article IEEE Transactions on Emerging Topics in Computational Intelligence · February 1, 2024 Automatic nuclei segmentation and classification play a vital role in digital pathology. However, previous works are mostly built on data with limited diversity and small sizes, making the results questionable or misleading in actual downstream tasks. In t ... Full text Cite

Scene Text Recognition via Dual-path Network with Shape-driven Attention Alignment

Journal article ACM Transactions on Multimedia Computing Communications and Applications · January 11, 2024 Scene text recognition (STR), one typical sequence-to-sequence problem, has drawn much attention recently in multimedia applications. To guarantee good performance, it is essential for STR to obtain aligned character-wise features from the whole-image feat ... Full text Cite

EPtask: Deep Reinforcement Learning Based Energy-Efficient and Priority-Aware Task Scheduling for Dynamic Vehicular Edge Computing

Journal article IEEE Transactions on Intelligent Vehicles · January 1, 2024 The increasing complexity of vehicles has led to a growing demand for in-vehicle services that rely on multiple sensors. In the Vehicular Edge Computing (VEC) paradigm, energy-efficient task scheduling is critical to achieving optimal completion time and e ... Full text Cite

Sim-to-Real Global Maximum Power Point Tracking With Domain Randomization and Adaptation for Photovoltaic Systems

Journal article IEEE Journal of Emerging and Selected Topics in Industrial Electronics · January 1, 2024 Simulations of photovoltaic (PV) systems help understand the nonlinear power-voltage characteristics in real-world atmospheric conditions. However, the gaps between simulation and real-world domain are usually significant due to the modeling errors. Theref ... Full text Cite

Semantic Similarity Distance: Towards better text-image consistency metric in text-to-image generation

Journal article Pattern Recognition · December 1, 2023 Generating high-quality images from text remains a challenge in visual-language understanding, with text-image consistency being a major concern. Particularly, the most popular metric R-precision may not accurately reflect the text-image consistency, leadi ... Full text Cite

FastAdaBelief: Improving Convergence Rate for Belief-Based Adaptive Optimizers by Exploiting Strong Convexity.

Journal article IEEE transactions on neural networks and learning systems · September 2023 AdaBelief, one of the current best optimizers, demonstrates superior generalization ability over the popular Adam algorithm by viewing the exponential moving average of observed gradients. AdaBelief is theoretically appealing in which it has a data-depende ... Full text Cite

Mind the Gap: Alleviating Local Imbalance for Unsupervised Cross-Modality Medical Image Segmentation.

Journal article IEEE journal of biomedical and health informatics · July 2023 Unsupervised cross-modality medical image adaptation aims to alleviate the severe domain gap between different imaging modalities without using the target domain label. A key in this campaign relies upon aligning the distributions of source and target doma ... Full text Cite

Explainable Tensorized Neural Ordinary Differential Equations for Arbitrary-Step Time Series Prediction

Journal article IEEE Transactions on Knowledge and Data Engineering · June 1, 2023 In this work, we propose a continuous neural network architecture, referred to as Explainable Tensorized Neural - Ordinary Differential Equations (ETN-ODE) network for multi-step time series prediction at arbitrary time points. Unlike existing approaches w ... Full text Cite

Towards Simple and Accurate Human Pose Estimation With Stair Network

Journal article IEEE Transactions on Emerging Topics in Computational Intelligence · June 1, 2023 In this paper, we focus on tackling the precise keypoint coordinates regression task. Most existing approaches adopt complicated networks with a large number of parameters, leading to a heavy model with poor cost-effectiveness in practice. To overcome this ... Full text Cite

Fitting Imbalanced Uncertainties in Multi-output Time Series Forecasting

Journal article ACM Transactions on Knowledge Discovery from Data · May 4, 2023 We focus on multi-step ahead time series forecasting with the multi-output strategy. From the perspective of multi-task learning (MTL), we recognize imbalanced uncertainties between prediction tasks of different future time steps. Unexpectedly, trained by ... Full text Cite

Randomized block-coordinate adaptive algorithms for nonconvex optimization problems

Journal article Engineering Applications of Artificial Intelligence · May 1, 2023 Nonconvex optimization problems have always been one focus in deep learning, in which many fast adaptive algorithms based on momentum are applied. However, the full gradient computation of high-dimensional feature vector in the above tasks become prohibiti ... Full text Cite

Generalized image outpainting with U-transformer.

Journal article Neural networks : the official journal of the International Neural Network Society · May 2023 In this paper, we develop a novel transformer-based generative adversarial neural network called U-Transformer for generalized image outpainting problems. Different from most present image outpainting methods conducting horizontal extrapolation, our genera ... Full text Cite

Towards Faster Training Algorithms Exploiting Bandit Sampling From Convex to Strongly Convex Conditions

Journal article IEEE Transactions on Emerging Topics in Computational Intelligence · April 1, 2023 The training process for deep learning and pattern recognition normally involves the use of convex and strongly convex optimization algorithms such as AdaBelief and SAdam to handle lots of 'uninformative' samples that should be ignored, thus incurring extr ... Full text Cite

Machine Learning Methods in Skin Disease Recognition: A Systematic Review

Journal article Processes · April 1, 2023 Skin lesions affect millions of people worldwide. They can be easily recognized based on their typically abnormal texture and color but are difficult to diagnose due to similar symptoms among certain types of lesions. The motivation for this study is to co ... Full text Cite

Rebalanced Zero-Shot Learning.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · January 2023 Zero-shot learning (ZSL) aims to identify unseen classes with zero samples during training. Broadly speaking, present ZSL methods usually adopt class-level semantic labels and compare them with instance-level semantic predictions to infer unseen classes. H ... Full text Cite

GSL-VO: A Geometric-Semantic Information Enhanced Lightweight Visual Odometry in Dynamic Environments

Journal article IEEE Transactions on Instrumentation and Measurement · January 1, 2023 Recently, learning-based visual odometry (VO) has attained remarkable success in vision-based measurement, especially in indoor robotics. Unfortunately, existing methods usually underexplore geometric-semantic (G-S) information, thus resulting in inefficie ... Full text Cite

Exploiting Attention-Consistency Loss For Spatial-Temporal Stream Action Recognition

Journal article ACM Transactions on Multimedia Computing Communications and Applications · October 6, 2022 Currently, many action recognition methods mostly consider the information from spatial streams. We propose a new perspective inspired by the human visual system to combine both spatial and temporal streams to measure their attention consistency. Specifica ... Full text Cite

A Novel 3D Unsupervised Domain Adaptation Framework for Cross-Modality Medical Image Segmentation.

Journal article IEEE journal of biomedical and health informatics · October 2022 We consider the problem of volumetric (3D) unsupervised domain adaptation (UDA) in cross-modality medical image segmentation, aiming to perform segmentation on the unannotated target domain (e.g. MRI) with the help of labeled source domain (e.g. CT). Previ ... Full text Cite

Zero-Shot Text Classification via Knowledge Graph Embedding for Social Media Data

Journal article IEEE Internet of Things Journal · June 15, 2022 The idea of 'citizen sensing' and 'human as sensors' is crucial for social Internet of Things, an integral part of cyber-physical-social systems (CPSSs). Social media data, which can be easily collected from the social world, has become a valuable resource ... Full text Cite

LightAdam: Towards a Fast and Accurate Adaptive Momentum Online Algorithm

Journal article Cognitive Computation · March 1, 2022 Adaptive optimization algorithms enjoy fast convergence and have been widely exploited in pattern recognition and cognitively-inspired machine learning. These algorithms may however be of high computational cost and low generalization ability due to their ... Full text Cite

Analyzing Cell-Scaffold Interaction through Unsupervised 3D Nuclei Segmentation.

Journal article International journal of bioprinting · January 2022 Fibrous scaffolds have been extensively used in three-dimensional (3D) cell culture systems to establish in vitro models in cell biology, tissue engineering, and drug screening. It is a common practice to characterize cell behaviors on such scaffold ... Full text Cite

Disentangling Semantic-to-Visual Confusion for Zero-Shot Learning

Journal article IEEE Transactions on Multimedia · January 1, 2022 Using generative models to synthesize visual features from semantic distribution is one of the most popular solutions to ZSL image classification in recent years. The triplet loss (TL) is popularly used to generate realistic visual distributions from seman ... Full text Cite

Artificial Intelligence in Collaborative Computing

Journal article Mobile Networks and Applications · December 1, 2021 Full text Cite

Coarse-grained generalized zero-shot learning with efficient self-focus mechanism

Journal article Neurocomputing · November 6, 2021 For image classification in computer vision, the performance of conventional deep neural networks (DNN) may usually drop when labeled training samples are limited. In this case, few-shot learning (FSL) or particularly zero-shot learning (ZSL), i.e. classif ... Full text Cite

Scaffold-A549: A Benchmark 3D Fluorescence Image Dataset for Unsupervised Nuclei Segmentation

Journal article Cognitive Computation · November 1, 2021 A general trend of nuclei segmentation is the transition from two-dimensional to three-dimensional nuclei segmentation and from traditional image processing methods to data-driven cognitively inspired methods. Existing nuclei segmentation datasets do not m ... Full text Cite

Improving generative adversarial networks with simple latent distributions

Journal article Neural Computing and Applications · October 1, 2021 Generative Adversarial Networks (GANs) have drawn great attention recently since they are the powerful models to generate high-quality images. Although GANs have achieved great success, they usually suffer from unstable training and consequently may lead t ... Full text Cite

Multi-modal generative adversarial networks for traffic event detection in smart cities

Journal article Expert Systems with Applications · September 1, 2021 Advances in the Internet of Things have enabled the development of many smart city applications and expert systems that help citizens and authorities better understand the dynamics of the cities, and make better planning and utilisation of city resources. ... Full text Cite

Manifold adversarial training for supervised and semi-supervised learning.

Journal article Neural networks : the official journal of the International Neural Network Society · August 2021 We propose a new regularization method for deep learning based on the manifold adversarial training (MAT). Unlike previous regularization and adversarial training methods, MAT further considers the local manifold of latent representations. Specifically, MA ... Full text Cite

Style-Neutralized Pattern Classification Based on Adversarially Trained Upgraded U-Net

Journal article Cognitive Computation · July 1, 2021 Traditional machine learning approaches usually hold the assumption that data for model training and in real applications are created following the identical and independent distribution (i.i.d.). However, several relevant research topics have demonstrated ... Full text Cite

Domain adaptation with feature and label adversarial networks

Journal article Neurocomputing · June 7, 2021 Learning a cross-domain representation from labeled source domains to unlabeled target domains is an important research problem in representation learning. Despite the success of traditional adversarial methods, they proposed to align features from each do ... Full text Cite

Automated Social Text Annotation With Joint Multilabel Attention Networks.

Journal article IEEE transactions on neural networks and learning systems · May 2021 Automated social text annotation is the task of suggesting a set of tags for shared documents on social media platforms. The automated annotation process can reduce users' cognitive overhead in tagging and improve tag management for better search, browsing ... Full text Cite

Attention-Augmented Machine Memory

Journal article Cognitive Computation · May 1, 2021 Attention mechanism plays an important role in the perception and cognition of human beings. Among others, many machine learning models have been developed to memorize the sequential data, such as the Long Short-Term Memory (LSTM) network and its extension ... Full text Cite

Novel Artificial Immune Networks-based optimization of shallow machine learning (ML) classifiers

Journal article Expert Systems with Applications · March 1, 2021 Artificial Immune Networks (AIN) is a population-based evolutionary algorithm that is inspired by theoretical immunology. It applies ideas and metaphors from the biological immune system to solve multi-disciplinary problems. This paper presents a novel app ... Full text Cite

State Primitive Learning to Overcome Catastrophic Forgetting in Robotics

Journal article Cognitive Computation · March 1, 2021 People can learn continuously a wide range of tasks without catastrophic forgetting. To mimic this functioning of continual learning, current methods mainly focus on studying a one-step supervised learning problem, e.g., image classification. They aim to r ... Full text Cite

A Multipath Fusion Strategy Based Single Shot Detector.

Journal article Sensors (Basel, Switzerland) · February 2021 Object detection has wide applications in intelligent systems and sensor applications. Compared with two stage detectors, recent one stage counterparts are capable of running more efficiently with comparable accuracy, which satisfy the requirement of real- ... Full text Cite

Residual attention-based multi-scale script identification in scene text images

Journal article Neurocomputing · January 15, 2021 Script identification is an essential step in the text extraction pipeline for multi-lingual application. This paper presents an effective approach to identify scripts in scene text images. Due to the complicated background, various text styles, character ... Full text Cite

A Systematic Analysis of Link Prediction in Complex Network

Journal article IEEE Access · January 1, 2021 Link mining is an important task in the field of data mining and has numerous applications in informal community. Suppose a real-world complex network, the responsibility of this function is to anticipate those links which are not occurred yet in the given ... Full text Cite

Neural CAPTCHA networks

Journal article Applied Soft Computing Journal · December 1, 2020 To protect against attacks by malicious computer programs, many websites apply the CAPTCHA (short for completely automated public turing test to tell computers and humans apart) technique for security protection. The distortion, rotation and deformation of ... Full text Cite

Generative adversarial classifier for handwriting characters super-resolution

Journal article Pattern Recognition · November 1, 2020 Generative Adversarial Networks (GAN) receive great attention recently due to its excellent performance in image generation, transformation, and super-resolution. However, less emphasis or study has been put on GAN for classification with super-resolution. ... Full text Cite

Compressing Deep Networks by Neuron Agglomerative Clustering.

Journal article Sensors (Basel, Switzerland) · October 2020 In recent years, deep learning models have achieved remarkable successes in various applications, such as pattern recognition, computer vision, and signal processing. However, high-performance deep architectures are often accompanied by a large storage spa ... Full text Cite

Improving deep neural network performance by integrating kernelized Min-Max objective

Journal article Neurocomputing · September 30, 2020 Deep neural networks (DNN), such as convolutional neural networks (CNN) have been widely used for object recognition. However, they are usually unable to ensure the required intra-class compactness and inter-class separability in the kernel space. These ar ... Full text Cite

Encoding primitives generation policy learning for robotic arm to overcome catastrophic forgetting in sequential multi-tasks learning.

Journal article Neural networks : the official journal of the International Neural Network Society · September 2020 Continual learning, a widespread ability in people and animals, aims to learn and acquire new knowledge and skills continuously. Catastrophic forgetting usually occurs in continual learning when an agent attempts to learn different tasks sequentially witho ... Full text Cite

Editorial: Collaborative Computing for Data-Driven Systems

Journal article Mobile Networks and Applications · August 1, 2020 Full text Cite

Segmentation mask guided end-to-end person search

Journal article Signal Processing Image Communication · August 1, 2020 Person search aims to search for a target person among multiple images recorded by multiple surveillance cameras, which faces various challenges from both pedestrian detection and person re-identification. Besides the large intra-class variations owing to ... Full text Cite

Triple loss for hard face detection

Journal article Neurocomputing · July 20, 2020 Although face detection has been well addressed in the last decades, despite the achievements in recent years, effective detection of small, blurred and partially occluded faces in the wild remains a challenging task. Meanwhile, the trade-off between compu ... Full text Cite

Novel deep neural network based pattern field classification architectures.

Journal article Neural networks : the official journal of the International Neural Network Society · July 2020 Field classification is a new extension of traditional classification frameworks that attempts to utilize consistent information from a group of samples (termed fields). By forgoing the independent identically distributed (i.i.d.) assumption, field classif ... Full text Cite

Generative adversarial networks with decoder-encoder output noises.

Journal article Neural networks : the official journal of the International Neural Network Society · July 2020 In recent years, research on image generation has been developing very fast. The generative adversarial network (GAN) emerges as a promising framework, which uses adversarial training to improve the generative ability of its generator. However, since GAN a ... Full text Cite

Knowledge base enrichment by relation learning from social tagging data

Journal article Information Sciences · July 1, 2020 There has been considerable interest in transforming unstructured social tagging data into structured knowledge for semantic-based retrieval and recommendation. Research in this line mostly exploits data co-occurrence and often overlooks the complex and am ... Full text Cite

Hybrid channel based pedestrian detection

Journal article Neurocomputing · May 14, 2020 Pedestrian detection has achieved great improvements with the help of Convolutional Neural Networks (CNNs). CNN can learn high-level features from input images, but the insufficient spatial resolution of CNN feature channels (feature maps) may cause a loss ... Full text Cite

Generative adversarial networks with mixture of t-distributions noise for diverse image generation.

Journal article Neural networks : the official journal of the International Neural Network Society · February 2020 Image generation is a long-standing problem in the machine learning and computer vision areas. In order to generate images with high diversity, we propose a novel model called generative adversarial networks with mixture of t-distributions noise (tGANs). I ... Full text Cite

Correlation Filter Selection for Visual Tracking Using Reinforcement Learning

Journal article IEEE Transactions on Circuits and Systems for Video Technology · January 1, 2020 Correlation filter has been proven to be an effective tool for a number of approaches in visual tracking, particularly for seeking a good balance between tracking accuracy and speed. However, correlation filter-based models are susceptible to wrong updates ... Full text Cite

Automatic Design of Deep Networks with Neural Blocks

Journal article Cognitive Computation · January 1, 2020 In recent years, deep neural networks (DNNs) have achieved great successes in many areas, such as cognitive computation, pattern recognition, and computer vision. Although many hand-crafted deep networks have been proposed in the literature, designing a we ... Full text Cite

A Novel Deep Density Model for Unsupervised Learning

Journal article Cognitive Computation · December 1, 2019 Density models are fundamental in machine learning and have received a widespread application in practical cognitive modeling tasks and learning problems. In this work, we introduce a novel deep density model, referred to as deep mixtures of factor analyze ... Full text Cite

Context-aware human activity and smartphone position-mining with motion sensors

Journal article Remote Sensing · November 1, 2019 Today's smartphones are equipped with embedded sensors, such as accelerometers and gyroscopes, which have enabled a variety of measurements and recognition tasks. In this paper, we jointly investigate two types of recognition problems in a joint manner, e. ... Full text Cite

Cross-modality interactive attention network for multispectral pedestrian detection

Journal article Information Fusion · October 1, 2019 Multispectral pedestrian detection is an emerging solution with great promise in many around-the-clock applications, such as automotive driving and security surveillance. To exploit the complementary nature and remedy contradictory appearance between modal ... Full text Cite

Discriminant Zero-Shot Learning with Center Loss

Journal article Cognitive Computation · August 15, 2019 Current work on zero-shot learning (ZSL) generally does not focus on the discriminative ability of the models, which is important for differentiating between classes since our brain focuses on the discriminating part of the object to classify it. For gener ... Full text Cite

Stochastic Conjugate Gradient Algorithm With Variance Reduction.

Journal article IEEE transactions on neural networks and learning systems · May 2019 Conjugate gradient (CG) methods are a class of important methods for solving linear equations and nonlinear optimization problems. In this paper, we propose a new stochastic CG algorithm with variance reduction1 and we prove its linear convergen ... Full text Cite

Special issue on advances in graph algorithm and applications

Journal article Neurocomputing · April 7, 2019 Full text Cite

Maximum Power Point Estimation for Photovoltaic Strings Subjected to Partial Shading Scenarios

Journal article IEEE Transactions on Industry Applications · March 1, 2019 Partial shading is an unavoidable complication in the field of photovoltaic (PV) generation. Bypass diodes have become a standard feature of solar cell arrays to improve array performance under partial shading scenarios (PSS). However, the current-voltage ... Full text Cite

IAN: The Individual Aggregation Network for Person Search

Journal article Pattern Recognition · March 1, 2019 Person search in real-world scenarios is a new challenging computer version task with many meaningful applications. The challenge of this task mainly comes from: (1) unavailable bounding boxes for pedestrians and the model needs to search for the person ov ... Full text Cite

Guided policy search for sequential multitask learning

Journal article IEEE Transactions on Systems Man and Cybernetics Systems · January 1, 2019 Policy search in reinforcement learning (RL) is a practical approach to interact directly with environments in parameter spaces, that often deal with dilemmas of local optima and real-time sample collection. A promising algorithm, known as guided policy se ... Full text Cite

Learning Latent Features with Infinite Nonnegative Binary Matrix Trifactorization

Journal article IEEE Transactions on Emerging Topics in Computational Intelligence · December 1, 2018 Nonnegative matrix factorization (NMF) has been widely exploited in many computational intelligence and pattern recognition problems. In particular, it can be used to extract latent features from data. However, previous NMF models often assume a fixed numb ... Full text Cite

Accelerating Infinite Ensemble of Clustering by Pivot Features

Journal article Cognitive Computation · December 1, 2018 The infinite ensemble clustering (IEC) incorporates both ensemble clustering and representation learning by fusing infinite basic partitions and shows appealing performance in the unsupervised context. However, it needs to solve the linear equation system ... Full text Cite

Three-Dimensional Local Energy-Based Shape Histogram (3D-LESH): A Novel Feature Extraction Technique

Journal article Expert Systems with Applications · December 1, 2018 In this paper, we present a novel feature extraction technique, termed Three-Dimensional Local Energy-Based Shape Histogram (3D-LESH), and exploit it to detect breast cancer in volumetric medical images. The technique is incorporated as part of an intellig ... Full text Cite

A new two-layer mixture of factor analyzers with joint factor loading model for the classification of small dataset problems

Journal article Neurocomputing · October 27, 2018 Dimensionality Reduction (DR) is a fundamental topic of pattern classification and machine learning. For classification tasks, DR is typically employed as a pre-processing step, succeeded by an independent classifier training stage. However, such independe ... Full text Cite

Banzhaf random forests: Cooperative game theory based random forests with consistency.

Journal article Neural networks : the official journal of the International Neural Network Society · October 2018 Random forests algorithms have been widely used in many classification and regression applications. However, the theory of random forests lags far behind their applications. In this paper, we propose a novel random forests classification algorithm based on ... Full text Cite

Approximately optimizing NDCG using pair-wise loss

Journal article Information Sciences · July 1, 2018 The Normalized Discounted Cumulative Gain (NDCG) is used to measure the performance of ranking algorithms. Much of the work on learning to rank by optimizing NDCG directly or indirectly is based on list-wise approaches. In our work, we approximately optimi ... Full text Cite

Reducing and Stretching Deep Convolutional Activation Features for Accurate Image Classification

Journal article Cognitive Computation · February 1, 2018 In order to extract effective representations of data using deep learning models, deep convolutional activation feature (DeCAF) is usually considered. However, since the deep models for learning DeCAF are generally pre-trained, the dimensionality of DeCAF ... Full text Cite

Learning from Few Samples with Memory Network

Journal article Cognitive Computation · February 1, 2018 Neural networks (NN) have achieved great successes in pattern recognition and machine learning. However, the success of a NN usually relies on the provision of a sufficiently large number of data samples as training data. When fed with a limited data set, ... Full text Cite

Special Issue Editorial: Cognitively-Inspired Computing for Knowledge Discovery

Journal article Cognitive Computation · February 1, 2018 Full text Cite

Zero-Shot Learning via Attribute Regression and Class Prototype Rectification.

Journal article IEEE transactions on image processing : a publication of the IEEE Signal Processing Society · February 2018 Zero-shot learning (ZSL) aims at classifying examples for unseen classes (with no training examples) given some other seen classes (with training examples). Most existing approaches exploit intermedia-level information (e.g., attributes) to transfer knowle ... Full text Cite

Siamese network ensemble for visual tracking

Journal article Neurocomputing · January 31, 2018 Visual object tracking is a challenging task considering illumination variation, occlusion, rotation, deformation and other problems. In this paper, we extend a Siamese INstance search Tracker (SINT) with model updating mechanism to improve its tracking ro ... Full text Cite

Novel Field-Support Vector Regression-Based Soft Sensor for Accurate Estimation of Solar Irradiance

Journal article IEEE Transactions on Circuits and Systems I Regular Papers · December 1, 2017 An accurate measurement of the solar irradiance is of significance for evaluating and developing of solar renewable energy systems. Soft sensors are used to provide feasible and economical alternatives to costly physical measurement instruments (e.g., pyra ... Full text Cite

Field Support Vector Machines

Journal article IEEE Transactions on Emerging Topics in Computational Intelligence · December 1, 2017 Conventional classifiers often regard input samples as identically and independently distributed (i.i.d.). This is however not true in many real applications, especially when patterns occur as groups (where each group shares a homogeneous style). Such task ... Full text Cite

Joint Learning of Unsupervised Dimensionality Reduction and Gaussian Mixture Model

Journal article Neural Processing Letters · June 1, 2017 Dimensionality reduction (DR) has been one central research topic in information theory, pattern recognition, and machine learning. Apparently, the performance of many learning models significantly rely on dimensionality reduction: successful DR can largel ... Full text Cite

Customer churn prediction in the telecommunication sector using a rough set approach

Journal article Neurocomputing · May 10, 2017 Customer churn is a critical and challenging problem affecting business and industry, in particular, the rapidly growing, highly competitive telecommunication sector. It is of substantial interest to both academic researchers and industrial practitioners, ... Full text Cite

A fast projected fixed-point algorithm for large graph matching

Journal article Pattern Recognition · December 1, 2016 We propose a fast algorithm for approximate matching of large graphs. Previous graph matching algorithms suffer from high computational complexity and therefore do not have good scalability. By using a new doubly stochastic projection, for matching two wei ... Full text Cite

Multicores and GPU utilization in parallel swarm algorithm for parameter estimation of photovoltaic cell model

Journal article Applied Soft Computing Journal · March 1, 2016 Bio-inspired metaheuristic algorithms have been widely applied in estimating the extrinsic parameters of a photovoltaic (PV) model. These methods are capable of handling the nonlinearity of objective functions whose derivatives are often not defined as wel ... Full text Cite

Maximum margin semi-supervised learning with irrelevant data.

Journal article Neural networks : the official journal of the International Neural Network Society · October 2015 Semi-supervised learning (SSL) is a typical learning paradigms training a model from both labeled and unlabeled data. The traditional SSL models usually assume unlabeled data are relevant to the labeled data, i.e., following the same distributions of the t ... Full text Cite

DE2: Dynamic ensemble of ensembles for learning nonstationary data

Journal article Neurocomputing · October 1, 2015 Learning nonstationary data with concept drift has received much attention in machine learning and been an active topic in ensemble learning. Specifically, batch growing ensemble methods present one important direction for dealing with concept drift involv ... Full text Cite

MTC: A Fast and Robust Graph-Based Transductive Learning Method.

Journal article IEEE transactions on neural networks and learning systems · September 2015 Despite the great success of graph-based transductive learning methods, most of them have serious problems in scalability and robustness. In this paper, we propose an efficient and robust graph-based transductive classification method, called minimum tree ... Full text Cite

Learning Imbalanced Classifiers Locally and Globally with One-Side Probability Machine

Journal article Neural Processing Letters · June 1, 2015 We consider the imbalanced learning problem, where the data associated with one class are far fewer than those associated with the other class. Current imbalanced learning methods often handle this problem by adapting certain intermediate parameters so as ... Full text Cite

Learning locality preserving graph from data.

Journal article IEEE transactions on cybernetics · November 2014 Machine learning based on graph representation, or manifold learning, has attracted great interest in recent years. As the discrete approximation of data manifold, the graph plays a crucial role in these kinds of learning approaches. In this paper, we prop ... Full text Cite

A novel classifier ensemble method with sparsity and diversity

Journal article Neurocomputing · June 25, 2014 We consider the classifier ensemble problem in this paper. Due to its superior performance to individual classifiers, class ensemble has been intensively studied in the literature. Generally speaking, there are two prevalent research directions on this, i. ... Full text Cite

Robust Text Detection in Natural Scene Images.

Journal article IEEE transactions on pattern analysis and machine intelligence · May 2014 Text detection in natural scene images is an important prerequisite for many content-based image analysis tasks. In this paper, we propose an accurate and robust method for detecting texts in natural scene images. A fast and effective pruning algorithm is ... Full text Cite

Graphical lasso quadratic discriminant function and its application to character recognition

Journal article Neurocomputing · April 10, 2014 Multivariate Gaussian distribution is a popular assumption in many pattern recognition tasks. The quadratic discriminant function (QDF) is an effective classification approach based on this assumption. An improved algorithm, called modified QDF (or MQDF in ... Full text Cite

Convex ensemble learning with sparsity and diversity

Journal article Information Fusion · January 1, 2014 Classifier ensemble has been broadly studied in two prevalent directions, i.e., to diversely generate classifier components, and to sparsely combine multiple classifiers. While most current approaches are emphasized on either sparsity or diversity only, we ... Full text Cite

Combination of classification and clustering results with label propagation

Journal article IEEE Signal Processing Letters · January 1, 2014 This letter considers the combination of multiple classification and clustering results to improve the prediction accuracy. First, an object-similarity graph is constructed from multiple clustering results. The labels predicted by the classification models ... Full text Cite

Geometry preserving multi-task metric learning

Journal article Machine Learning · July 1, 2013 In this paper, we consider the multi-task metric learning problem, i.e., the problem of learning multiple metrics from several correlated tasks simultaneously. Despite the importance, there are only a limited number of approaches in this field. While the e ... Full text Cite

A multi-task framework for metric learning with common subspace

Journal article Neural Computing and Applications · June 1, 2013 Metric learning has been widely studied in machine learning due to its capability to improve the performance of various algorithms. Meanwhile, multi-task learning usually leads to better performance by exploiting the shared information across all tasks. In ... Full text Cite

Maxi-Min discriminant analysis via online learning.

Journal article Neural networks : the official journal of the International Neural Network Society · October 2012 Linear Discriminant Analysis (LDA) is an important dimensionality reduction algorithm, but its performance is usually limited on multi-class data. Such limitation is incurred by the fact that LDA actually maximizes the average divergence among classes, whe ... Full text Cite

Joint learning of error-correcting output codes and dichotomizers from data

Journal article Neural Computing and Applications · June 1, 2012 The ECOC technique is a powerful tool to learn and combine multiple binary learners for multi-class classification. It generally involves three steps: coding, dichotomizers learning, and decoding. In previous ECOC methods, the coding step and the dichotomi ... Full text Cite

FMI image based rock structure classification using classifier combination

Journal article Neural Computing and Applications · October 1, 2011 Formation Micro Imager (FMI) can directly reflect changes of wall stratums and rock structures, and is an important factor to classify stratums and identify lithology for the oil and gas exploration. Conventionally, people analyze FMI images mainly with ma ... Full text Cite

Exchange rate prediction with non-numerical information

Journal article Neural Computing and Applications · October 1, 2011 Exchange rate prediction is an important yet challenging problem in financial time series analysis. Although the historical exchange rates can provide valuable information, other factors will also affect the prediction significantly. These factors could be ... Full text Cite

Generalized sparse metric learning with relative comparisons

Journal article Knowledge and Information Systems · January 1, 2011 The objective of sparse metric learning is to learn a distance measure from a set of data in addition to finding a low-dimensional representation. Despite demonstrated success, the performance of existing sparse metric learning approaches is usually limite ... Full text Cite

Sparse learning for support vector classification

Journal article Pattern Recognition Letters · October 1, 2010 This paper provides a sparse learning algorithm for Support Vector Classification (SVC), called Sparse Support Vector Classification (SSVC), which leads to sparse solutions by automatically setting the irrelevant parameters exactly to zero. SSVC adopts the ... Full text Cite

A novel kernel-based maximum a posteriori classification method.

Journal article Neural networks : the official journal of the International Neural Network Society · September 2009 Kernel methods have been widely used in pattern recognition. Many kernel classifiers such as Support Vector Machines (SVM) assume that data can be separated by a hyperplane in the kernel-induced feature space. These methods do not consider the data distrib ... Full text Cite

Enhanced protein fold recognition through a novel data integration approach.

Journal article BMC bioinformatics · August 2009 BackgroundProtein fold recognition is a key step in protein three-dimensional (3D) structure discovery. There are multiple fold discriminatory data sources which use physicochemical and structural properties as well as further data sources derived ... Full text Cite

Localized support vector regression for time series prediction

Journal article Neurocomputing · June 1, 2009 Time series prediction, especially financial time series prediction, is a challenging task in machine learning. In this issue, the data are usually non-stationary and volatile in nature. Because of its good generalization power, the support vector regressi ... Full text Cite

Arbitrary norm support vector machines.

Journal article Neural computation · February 2009 Support vector machines (SVM) are state-of-the-art classifiers. Typically L2-norm or L1-norm is adopted as a regularization term in SVMs, while other norm-based SVMs, for example, the L0-norm SVM or even the L(infinity)-norm SVM, are rarely seen in the lit ... Full text Cite

Maxi-min margin machine: learning large margin classifiers locally and globally.

Journal article IEEE transactions on neural networks · February 2008 In this paper, we propose a novel large margin classifier, called the maxi-min margin machine M(4). This model learns the decision boundary both locally and globally. In comparison, other large margin classifiers construct separating hyperplanes only eithe ... Full text Cite

Imbalanced learning with a biased minimax probability machine.

Journal article IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society · August 2006 Imbalanced learning is a challenged task in machine learning. In this context, the data associated with one class are far fewer than those associated with the other class. Traditional machine learning methods seeking classification accuracy over a full ran ... Full text Cite

Maximizing sensitivity in medical diagnosis using biased minimax probability machine.

Journal article IEEE transactions on bio-medical engineering · May 2006 The challenging task of medical diagnosis based on machine learning techniques requires an inherent bias, i.e., the diagnosis should favor the "ill" class over the "healthy" class, since misdiagnosing a patient as a healthy person may delay the therapy and ... Full text Cite

The minimum error minimax probability machine

Journal article Journal of Machine Learning Research · October 1, 2004 We construct a distribution-free Bayes optimal classifier called the Minimum Error Minimax Probability Machine (MEMPM) in a worst-case setting, i.e., under all possible choices of class-conditional densities with a given mean and covariance matrix. By assu ... Cite