David B. Dunson CV

Conference Proceedings of SPIE the International Society for Optical Engineering · September 17, 2025 Alzheimer’s disease (AD) affects over 10% of people above age 65. Current treatments remain largely ineffective, thus early biomarkers are essential for devising preventive interventions, and personalizing these based on risk profiles. Brain age gap (BAG)— ... Full text Cite

Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI

Conference Proceedings of Machine Learning Research · January 1, 2024 In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metr ... Cite

Statistical Guarantees for Transformation Based Models with Applications to Implicit Variational Inference

Conference Proceedings of Machine Learning Research · January 1, 2021 Transformation-based methods have been an attractive approach in non-parametric inference for problems such as unconditional and conditional density estimation due to their unique hierarchical structure that models the data as flexible transformation of a ... Cite

Fiedler regularization: Learning neural networks with graph sparsity

Conference 37th International Conference on Machine Learning Icml 2020 · January 1, 2020 We introduce a novel regularization approach for deep learning that incorporates and respects the underlying graphical structure of the neural network. Existing regularization methods often focus on penalizing weights in a global/uniform manner that ignore ... Cite

Fiedler Regularization: Learning Neural Networks with Graph Sparsity

Conference Proceedings of Machine Learning Research · January 1, 2020 We introduce a novel regularization approach for deep learning that incorporates and respects the underlying graphical structure of the neural network. Existing regularization methods often focus on penalizing weights in a global/uniform manner that ignore ... Cite

Locally convex kernel mixtures: Bayesian subspace learning

Conference Proceedings 18th IEEE International Conference on Machine Learning and Applications Icmla 2019 · December 1, 2019 Kernel mixture models are routinely used for density estimation. However, in multivariate settings, issues arise in efficiently approximating lower-dimensional structure in the data. For example, it is common to suppose that the density is concentrated nea ... Full text Cite

Effect of A1C and Glucose on Postoperative Mortality in Noncardiac and Cardiac Surgeries.

Conference Diabetes Care · April 2018 OBJECTIVE: Hemoglobin A1c (A1C) is used in assessment of patients for elective surgeries because hyperglycemia increases risk of adverse events. However, the interplay of A1C, glucose, and surgical outcomes remains unclarified, with often only two of these ... Full text Link to item Cite

No penalty no tears: Least squares in high-dimensional linear models

Conference 33rd International Conference on Machine Learning Icml 2016 · January 1, 2016 Ordinary least squares (OI,S) is the default method for fitting linear models, but is not applicable for problems with dimensionality larger than the sample size. For these problems, we advocate the use of a generalized version of OLS motivated by ridge re ... Cite

DECOrrelated feature space partitioning for distributed sparse regression

Conference Advances in Neural Information Processing Systems · January 1, 2016 Fitting statistical models is computationally challenging when the sample size or the dimension of the dataset is huge. An attractive approach for down-scaling the problem size is to first partition the dataset into subsets and then fit using distributed a ... Cite

Scalable geometric density estimation

Conference Proceedings of the 19th International Conference on Artificial Intelligence and Statistics Aistats 2016 · January 1, 2016 It is standard to assume a low-dimensional structure in estimating a high-dimensional density. However, popular methods, such as probabilistic principal component analysis, scale poorly computationally. We introduce a novel empirical Bayes method that we t ... Cite

Variational Gaussian copula inference

Conference Proceedings of the 19th International Conference on Artificial Intelligence and Statistics Aistats 2016 · January 1, 2016 We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models. For models with continuous and non-Gaussian hidden variables, we propose a semiparametric and automated variational ... Cite

Uncovering systematic bias in ratings across categories: A Bayesian approach

Conference Recsys 2015 Proceedings of the 9th ACM Conference on Recommender Systems · September 16, 2015 Recommender systems are routinely equipped with standardized taxonomy that associates each item with one or more categories or genres. Although such information does not directly imply the quality of an item, the distribution of ratings vary greatly across ... Full text Cite

WASP: Scalable Bayes via barycenters of subset posteriors

Conference Journal of Machine Learning Research · January 1, 2015 The promise of Bayesian methods for big data sets has not fully been realized due to the lack of scalable computational algorithms. For massive data, it is necessary to store and process subsets on different machines in a distributed manner. We propose a s ... Cite

Quantifying uncertainty in variable selection with arbitrary matrices

Conference 2015 IEEE 6th International Workshop on Computational Advances in Multi Sensor Adaptive Processing Camsap 2015 · January 1, 2015 Probabilistically quantifying uncertainty in parameters, predictions and decisions is a crucial component of broad scientific and engineering applications. This is however difficult if the number of parameters far exceeds the sample size. Although there ar ... Full text Cite

On the consistency theory of high dimensional variable screening

Conference Advances in Neural Information Processing Systems · January 1, 2015 Variable screening is a fast dimension reduction technique for assisting high dimensional feature selection. As a preselection method, it selects a moderate size subset of candidate variables for further refining via feature selection to produce the final ... Cite

Parallelizing MCMC with random partition trees

Conference Advances in Neural Information Processing Systems · January 1, 2015 The modern scale of data has brought new challenges to Bayesian inference. In particular, conventional MCMC algorithms are computationally very expensive for large data sets. A promising approach to solve this problem is embarrassingly parallel MCMC (EP-MC ... Cite

Probabilistic curve learning: Coulomb repulsion and the electrostatic Gaussian process

Conference Advances in Neural Information Processing Systems · January 1, 2015 Learning of low dimensional structure in multidimensional data is a canonical problem in machine learning. One common approach is to suppose that the observed data are close to a lower-dimensional smooth manifold. There are a rich variety of manifold learn ... Cite

Digital cradle removal in X-ray images of art paintings

Conference 2014 IEEE International Conference on Image Processing Icip 2014 · January 28, 2014 We introduce an algorithm that removes the deleterious effect of cradling on X-ray images of paintings on wooden panels. The algorithm consists of a three stage procedure. Firstly, the cradled regions are located automatically. The second step consists of ... Full text Cite

Scalable bayesian low-rank decomposition of incomplete multiway tensors

Conference 31st International Conference on Machine Learning Icml 2014 · January 1, 2014 We present a scalable Bayesian framework for low-rank decomposition of multiway tensor data with missing observations. The key issue of pre-specifying the rank of the decomposition is sidestepped in a principled manner using a multiplicative gamma process ... Cite

Scalable and robust Bayesian inference via the median posterior

Conference 31st International Conference on Machine Learning Icml 2014 · January 1, 2014 Many Bayesian learning methods for massive data benefit from working with small subsets of observations. In particular, significant progress has been made in scalable Bayesian learning via stochastic approximation. However, Bayesian learning methods in dis ... Cite

Median selection subset aggregation for parallel inference

Conference Advances in Neural Information Processing Systems · January 1, 2014 For massive data sets, efficient computation commonly relies on distributed algorithms that store and process subsets of the data on different machines, minimizing communication costs. Our focus is on regression and classification problems involving many f ... Cite

Bayesian logistic Gaussian process models for dynamic networks

Conference Journal of Machine Learning Research · January 1, 2014 Time-varying adjacency matrices encoding the presence or absence of a relation among entities are available in many research fields. Motivated by an application to studying dynamic networks among sports teams, we propose a Bayesian nonparametric model. The ... Cite

Diagonal orthant multinomial probit models

Conference Journal of Machine Learning Research · January 1, 2013 Bayesian classification commonly relies on probit models, with data augmentation algorithms used for posterior computation. By imputing latent Gaussian variables, one can often trivially adapt computational approaches used in Gaussian models. However, MCMC ... Cite

Bayesian learning of joint distributions of objects

Conference Journal of Machine Learning Research · January 1, 2013 There is increasing interest in broad application areas in defining flexible joint models for data having a variety of measurement scales, while also allowing data of complex types, such as functions, images and documents. We consider a general framework f ... Cite

Hierarchical latent dictionaries for models of brain activation

Conference Journal of Machine Learning Research · January 1, 2012 In this work, we propose a hierarchical latent dictionary approach to estimate the timevarying mean and covariance of a process for which we have only limited noisy samples. We fully leverage the limited sample size and redundancy in sensor measurements by ... Cite

The kernel beta process

Conference Advances in Neural Information Processing Systems 24 25th Annual Conference on Neural Information Processing Systems 2011 Nips 2011 · January 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

Conference Advances in Neural Information Processing Systems 24 25th Annual Conference on Neural Information Processing Systems 2011 Nips 2011 · January 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

Generalized beta mixtures of Gaussians

Conference Advances in Neural Information Processing Systems 24 25th Annual Conference on Neural Information Processing Systems 2011 Nips 2011 · January 1, 2011 In recent years, a rich variety of shrinkage priors have been proposed that have great promise in addressing massive regression problems. In general, these new priors can be expressed as scale mixtures of normals, but have more complex forms and better pro ... Cite

Joint analysis of time-evolving binary matrices and associated documents

Conference Advances in Neural Information Processing Systems 23 24th Annual Conference on Neural Information Processing Systems 2010 Nips 2010 · January 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