1 code implementation • 10 Feb 2024 • Changwoo J. Lee, Alessandro Zito, Huiyan Sang, David B. Dunson
The beta distribution serves as a canonical tool for modeling probabilities and is extensively used in statistics and machine learning, especially in the field of Bayesian nonparametrics.
no code implementations • 20 Dec 2023 • Piotr M. Suder, Jason Xu, David B. Dunson
Transfer learning is a burgeoning concept in statistical machine learning that seeks to improve inference and/or predictive accuracy on a domain of interest by leveraging data from related domains.
no code implementations • 24 Nov 2023 • Duy H. Thai, Alexander L. Young, David B. Dunson
We develop a novel class of MCMC algorithms based on a stochastized Nesterov scheme.
no code implementations • 19 Oct 2023 • Steven Winter, Omar Melikechi, David B. Dunson
Gibbs posteriors are proportional to a prior distribution multiplied by an exponentiated loss function, with a key tuning parameter weighting information in the loss relative to the prior and providing a control of posterior uncertainty.
no code implementations • 21 Apr 2023 • Steven Winter, Trevor Campbell, Lizhen Lin, Sanvesh Srivastava, David B. Dunson
Bayesian models are a powerful tool for studying complex data, allowing the analyst to encode rich hierarchical dependencies and leverage prior information.
2 code implementations • 20 Apr 2023 • Omar Melikechi, David B. Dunson
We introduce Cayley transform ellipsoid fitting (CTEF), an algorithm that uses the Cayley transform to fit ellipsoids to noisy data in any dimension.
no code implementations • 9 Mar 2021 • Georgia Papadogeorgou, Carolina Bello, Otso Ovaskainen, David B. Dunson
Reductions in natural habitats urge that we better understand species' interconnection and how biological communities respond to environmental changes.
Methodology
no code implementations • 18 Oct 2020 • Lizhen Lin, Bayan Saparbayeva, Michael Minyi Zhang, David B. Dunson
One of the key challenges for optimization on manifolds is the difficulty of verifying the complexity of the objective function, e. g., whether the objective function is convex or non-convex, and the degree of non-convexity.
1 code implementation • 16 Jul 2020 • Sirio Legramanti, Tommaso Rigon, Daniele Durante, David B. Dunson
The coexistence of these noisy block patterns limits the reliability of routinely-used community detection algorithms, and requires extensions of model-based solutions to realistically characterize the node partition process, incorporate information from node attributes, and provide improved strategies for estimation and uncertainty quantification.
no code implementations • 9 Jun 2020 • Tommaso Rigon, Amy H. Herring, David B. Dunson
Several existing clustering algorithms, including k-means, can be interpreted as generalized Bayes estimators under our framework, and hence we provide a method of uncertainty quantification for these approaches.
no code implementations • 26 Apr 2020 • Yichen Zhu, Cheng Li, David B. Dunson
When data are limited in one or more of the classes, the estimated decision boundaries are often irregularly shaped due to the limited sample size, leading to poor generalization error.
no code implementations • 3 Apr 2020 • Emanuele Aliverti, David B. Dunson
Multivariate categorical data are routinely collected in many application areas.
Methodology
1 code implementation • 17 Mar 2020 • Shounak Chattopadhyay, Antik Chakraborty, David B. Dunson
There is a rich literature on Bayesian methods for density estimation, which characterize the unknown density as a mixture of kernels.
no code implementations • 12 Jan 2020 • Meimei Liu, David B. Dunson
When there is a distributional shift between data used to train a predictive algorithm and current data, performance can suffer.
no code implementations • 7 Nov 2019 • Meimei Liu, Zhengwu Zhang, David B. Dunson
In this paper, building on recent advances in deep learning, we develop a nonlinear latent factor model to characterize the population distribution of brain graphs and infer the relationships between brain structural connectomes and human traits.
1 code implementation • 5 Nov 2019 • Federico Ferrari, David B. Dunson
Our interest is in model selection for these different components, accounting for uncertainty and addressing non-identifability between the linear and nonparametric components of the semiparametric model.
Applications
1 code implementation • 7 Oct 2019 • Arkaprava Roy, Isaac Lavine, Amy H. Herring, David B. Dunson
As an alternative, we propose a class of Perturbed Factor Analysis (PFA) models that assume a common factor structure across studies after perturbing the data via multiplication by a study-specific matrix.
Methodology
1 code implementation • 15 Sep 2019 • Willem van den Boom, Galen Reeves, David B. Dunson
Posterior computation for high-dimensional data with many parameters can be challenging.
Computation Methodology
no code implementations • 29 Jun 2019 • Didong Li, David B. Dunson
When the manifold is unknown, it is challenging to accurately approximate the geodesic distance.
no code implementations • 25 Apr 2019 • Federico Ferrari, David B. Dunson
This article is motivated by the problem of inference on interactions among chemical exposures impacting human health outcomes.
Methodology Applications
no code implementations • 25 Apr 2019 • Xu Zhu, David B. Dunson
We consider the Lipschitz bandit optimization problem with an emphasis on practical efficiency.
1 code implementation • 3 Mar 2019 • Didong Li, David B. Dunson
It is challenging to obtain accurate classification performance when the feature distributions in the different classes are complex, with nonlinear, overlapping and intersecting supports.
1 code implementation • 12 Feb 2019 • Sirio Legramanti, Daniele Durante, David B. Dunson
There is a wide variety of models in which the dimension of the parameter space is unknown.
Methodology
no code implementations • 1 Jan 2019 • Shaobo Han, David B. Dunson
We introduce a multiscale supervised dimension reduction method for SPatial Interaction Network (SPIN) data, which consist of a collection of spatially coordinated interactions.
no code implementations • 19 Oct 2018 • Leo L. Duan, David B. Dunson
Model-based clustering is widely-used in a variety of application areas.
1 code implementation • 16 Sep 2018 • Michele Peruzzi, David B. Dunson
We tackle the problem of multiscale regression for predictors that are spatially or temporally indexed, or with a pre-specified multiscale structure, with a Bayesian modular approach.
Methodology
no code implementations • 3 Mar 2018 • Shaobo Han, David B. Dunson
This article is motivated by soccer positional passing networks collected across multiple games.
no code implementations • 28 Sep 2017 • Ivo D. Shterev, David B. Dunson, Cliburn Chan, Gregory D. Sempowski
High throughput screening of compounds (chemicals) is an essential part of drug discovery [7], involving thousands to millions of compounds, with the purpose of identifying candidate hits.
3 code implementations • 26 Jun 2017 • Didong Li, Minerva Mukhopadhyay, David B. Dunson
There is a rich literature on approximating the unknown manifold, and on exploiting such approximations in clustering, data compression, and prediction.
no code implementations • 17 Nov 2016 • Fangjian Guo, Xiangyu Wang, Kai Fan, Tamara Broderick, David B. Dunson
Variational inference (VI) provides fast approximations of a Bayesian posterior in part because it formulates posterior approximation as an optimization problem: to find the closest distribution to the exact posterior over some family of distributions.
no code implementations • 17 Mar 2016 • Shiwen Zhao, Barbara E. Engelhardt, Sayan Mukherjee, David B. Dunson
We illustrate the utility of our approach on simulated data, comparing results from MELD to alternative methods, and we show the promise of our approach through the application of MELD to several data sets.
1 code implementation • 19 Jun 2015 • Shaobo Han, Xuejun Liao, David B. Dunson, Lawrence Carin
We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models.
no code implementations • NeurIPS 2015 • Ye Wang, David B. Dunson
Learning of low dimensional structure in multidimensional data is a canonical problem in machine learning.
2 code implementations • NeurIPS 2015 • Xiangyu Wang, Fangjian Guo, Katherine A. Heller, David B. Dunson
The new algorithm applies random partition trees to combine the subset posterior draws, which is distribution-free, easy to resample from and can adapt to multiple scales.
no code implementations • 21 May 2015 • Daniele Durante, David B. Dunson
Our focus is on realistically modeling and forecasting dynamic networks of face-to-face contacts among individuals.
no code implementations • NeurIPS 2015 • Xiangyu Wang, Chenlei Leng, David B. Dunson
Variable screening is a fast dimension reduction technique for assisting high dimensional feature selection.
no code implementations • 21 Jan 2015 • Yan Shang, David B. Dunson, Jing-Sheng Song
In cargo logistics, a key performance measure is transport risk, defined as the deviation of the actual arrival time from the planned arrival time.
no code implementations • 7 Jun 2014 • Rajarshi Guhaniyogi, David B. Dunson
Nonparametric regression for massive numbers of samples (n) and features (p) is an increasingly important problem.
no code implementations • 11 Mar 2014 • Stanislav Minsker, Sanvesh Srivastava, Lizhen Lin, David B. Dunson
We propose a novel approach to Bayesian analysis that is provably robust to outliers in the data and often has computational advantages over standard methods.
no code implementations • 6 Mar 2014 • Yun Yang, David B. Dunson
It is generally believed that ensemble approaches, which combine multiple algorithms or models, can outperform any single algorithm at machine learning tasks, such as prediction.
1 code implementation • 15 Jan 2014 • Shaan Qamar, Rajarshi Guhaniyogi, David B. Dunson
We propose a Conditional Density Filtering (C-DF) algorithm for efficient online Bayesian inference.
1 code implementation • 17 Dec 2013 • Xiangyu Wang, David B. Dunson
With the rapidly growing scales of statistical problems, subset based communication-free parallel MCMC methods are a promising future for large scale Bayesian analysis.
no code implementations • NeurIPS 2013 • Francesca Petralia, Joshua Vogelstein, David B. Dunson
Nonparametric estimation of the conditional distribution of a response given high-dimensional features is a challenging problem.
no code implementations • NeurIPS 2013 • Daniele Durante, Bruno Scarpa, David B. Dunson
In modeling multivariate time series, it is important to allow time-varying smoothness in the mean and covariance process.
no code implementations • 19 Nov 2013 • Daniele Durante, David B. Dunson
Symmetric binary matrices representing relations among entities are commonly collected in many areas.
no code implementations • 26 Apr 2013 • David C. Kessler, Jack Taylor, David B. Dunson
Learning a distribution conditional on a set of discrete-valued features is a commonly encountered task.
no code implementations • 4 Mar 2013 • Rajarshi Guhaniyogi, David B. Dunson
As an alternative to variable selection or shrinkage in high dimensional regression, we propose to randomly compress the predictors prior to analysis.
no code implementations • NeurIPS 2012 • Emily Fox, David B. Dunson
We propose a multiresolution Gaussian process to capture long-range, non-Markovian dependencies while allowing for abrupt changes.
no code implementations • NeurIPS 2012 • Francesca Petralia, Vinayak Rao, David B. Dunson
One important issue that arises in using discrete mixtures is low separation in the components; in particular, different components can be introduced that are very similar and hence redundant.
no code implementations • 7 Oct 2012 • Daniele Durante, Bruno Scarpa, David B. Dunson
In modeling multivariate time series, it is important to allow time-varying smoothness in the mean and covariance process.
no code implementations • NeurIPS 2011 • Artin Armagan, Merlise Clyde, David B. Dunson
In recent years, a rich variety of shrinkage priors have been proposed that have great promise in addressing massive regression problems.
no code implementations • NeurIPS 2011 • Xianxing Zhang, Lawrence Carin, David B. Dunson
The nested Chinese restaurant process is extended to design a nonparametric topic-model tree for representation of human choices.
no code implementations • NeurIPS 2011 • Lu Ren, Yingjian Wang, Lawrence Carin, David B. Dunson
A new Le ́vy process prior is proposed for an uncountable collection of covariate- dependent feature-learning measures; the model is called the kernel beta process (KBP).
no code implementations • NeurIPS 2010 • Eric Wang, Dehong Liu, Jorge Silva, Lawrence Carin, David B. Dunson
An objective of such analysis is to infer structure and inter-relationships underlying the matrices, here defined by latent features associated with each axis of the matrix.
no code implementations • NeurIPS 2009 • Lan Du, Lu Ren, Lawrence Carin, David B. Dunson
The model clusters the images into classes, and each image is segmented into a set of objects, also allowing the opportunity to assign a word to each object (localized labeling).