Dirichlet-Laplace priors for optimal shrinkage

DOI10.1080/01621459.2014.960967MaRDI QIDQ5367461zbMATH OpenOpenAlexWikidataFDO

Authors Anirban Bhattacharya, Debdeep Pati, Natesh S. Pillai, David Dunson

Publication date 13 October 2017

Published in Journal of the American Statistical Association (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1401.5398

zbMATH Keywords

LassoBayesianregularizationhigh-dimensionalpenalized regressionconvergence rateshrinkage priorDirichlet-Laplace\(L_1\)

Mathematics Subject Classification ID

Bayesian inference (62F15) Ridge regression; shrinkage estimators (Lasso) (62J07)

Abstract: Penalized regression methods, such as $L_1$ regularization, are routinely used in high-dimensional applications, and there is a rich literature on optimality properties under sparsity assumptions. In the Bayesian paradigm, sparsity is routinely induced through two-component mixture priors having a probability mass at zero, but such priors encounter daunting computational problems in high dimensions. This has motivated an amazing variety of continuous shrinkage priors, which can be expressed as global-local scale mixtures of Gaussians, facilitating computation. In sharp contrast to the frequentist literature, little is known about the properties of such priors and the convergence and concentration of the corresponding posterior distribution. In this article, we propose a new class of Dirichlet--Laplace (DL) priors, which possess optimal posterior concentration and lead to efficient posterior computation exploiting results from normalized random measure theory. Finite sample performance of Dirichlet--Laplace priors relative to alternatives is assessed in simulated and real data examples.

Recommendations

• Sub-optimality of some continuous shrinkage priors
• Contraction properties of shrinkage priors in logistic regression
• Bayesian Regression Using a Prior on the Model Fit: The R2-D2 Shrinkage Prior
• Generalized double Pareto shrinkage
• Variable selection via penalized credible regions with Dirichlet-Laplace global-local shrinkage priors

Cited in

(only showing first 100 items - show all)

• Precision matrix estimation under the horseshoe-like prior-penalty dual
• APPROXIMATE BAYESIAN INFERENCE AND FORECASTING IN HUGE‐DIMENSIONAL MULTICOUNTRY VARs
• Bayesian semiparametric functional mixed models for serially correlated functional data, with application to glaucoma data
• Bayesian inference in nonparanormal graphical models
• Singular value shrinkage priors for Bayesian prediction
• Computationally efficient inference in large Bayesian mixed frequency VARs
• Dealing with Measurement Uncertainties as Nuisance Parameters in Bayesian Model Calibration
• Ultra high-dimensional multivariate posterior contraction rate under shrinkage priors
• Sparsity information and regularization in the horseshoe and other shrinkage priors
• Hierarchical shrinkage Gaussian processes: applications to computer code emulation and dynamical system recovery
• Dynamic shrinkage priors for large time-varying parameter regressions using scalable Markov chain Monte Carlo methods
• Regression with variable dimension covariates
• Bayesian Approaches to Shrinkage and Sparse Estimation
• Functional Horseshoe Priors for Subspace Shrinkage
• Group-wise shrinkage estimation in penalized model-based clustering
• Testing Sparsity-Inducing Penalties
• Targeted random projection for prediction from high-dimensional features
• A novel Bayesian approach for variable selection in linear regression models
• Bayesian analysis for matrix-variate logistic regression with/without response misclassification
• A Bayesian panel vector autoregression to analyze the impact of climate shocks on high-income economies
• Relevant parameter changes in structural break models
• Bayesian Bootstrap Spike-and-Slab LASSO
• Distributed Bayesian posterior voting strategy for massive data
• Variable selection via penalized credible regions with Dirichlet-Laplace global-local shrinkage priors
• The illusion of the illusion of sparsity: an exercise in prior sensitivity
• Large Bayesian SVARs with linear restrictions
• Spike-and-Slab Group Lassos for Grouped Regression and Sparse Generalized Additive Models
• MIMIX: a Bayesian mixed-effects model for microbiome data from designed experiments
• Modeling time-varying parameters using artificial neural networks: a GARCH illustration
• Bayesian shrinkage in mixture-of-experts models: identifying robust determinants of class membership
• On minimax optimality of sparse Bayes predictive density estimates
• Shared Bayesian variable shrinkage in multinomial logistic regression
• Fast Monte Carlo Markov chains for Bayesian shrinkage models with random effects
• Variable selection using shrinkage priors
• Uncertainty Quantification for Modern High-Dimensional Regression via Scalable Bayesian Methods
• Bayesian sparse spiked covariance model with a continuous matrix shrinkage prior
• Informative Bayesian neural network priors for weak signals
• Global-local shrinkage priors for asymptotic point and interval estimation of normal means under sparsity
• Prior-Preconditioned Conjugate Gradient Method for Accelerated Gibbs Sampling in “Large n , Large p ” Bayesian Sparse Regression
• Bayesian hierarchical modeling on covariance valued data
• The Graphical Horseshoe Estimator for Inverse Covariance Matrices
• Empirical priors for prediction in sparse high-dimensional linear regression
• Data-driven priors and their posterior concentration rates
• Trace class Markov chains for Bayesian inference with generalized double Pareto shrinkage priors
• Prediction risk for the horseshoe regression
• The Bayesian group Lasso for confounded spatial data
• A robust Bayesian analysis of variable selection under prior ignorance
• Adaptive Bayesian nonparametric regression using a kernel mixture of polynomials with application to partial linear models
• Structured Shrinkage Priors
• Neuronized Priors for Bayesian Sparse Linear Regression
• Optimal false discovery control of minimax estimators
• Model selection using mass-nonlocal prior
• Fast Markov Chain Monte Carlo for High-Dimensional Bayesian Regression Models With Shrinkage Priors
• Penalising model component complexity: a principled, practical approach to constructing priors
• Radio-iBAG: radiomics-based integrative Bayesian analysis of multiplatform genomic data
• Spike and slab empirical Bayes sparse credible sets
• Shrinkage priors for Bayesian penalized regression
• Bayesian modeling of interaction between features in sparse multivariate count data with application to microbiome study
• A horseshoe mixture model for Bayesian screening with an application to light sheet fluorescence microscopy in brain imaging
• Generalized double Pareto shrinkage
• Bayesian regression trees for high-dimensional prediction and variable selection
• Bayesian approaches to variable selection: a comparative study from practical perspectives
• Bayesian wavelet shrinkage with logistic prior
• Asymmetric Prior in Wavelet Shrinkage
• Compound Poisson processes, latent shrinkage priors and Bayesian nonconvex penalization
• Conditions for posterior contraction in the sparse normal means problem
• Predictive performance of Dirichlet process shrinkage methods in linear regression
• Wavelet shrinkage in nonparametric regression models with positive noise
• Projective inference in high-dimensional problems: prediction and feature selection
• Sub-optimality of some continuous shrinkage priors
• Efficient Sampling for Gaussian Linear Regression With Arbitrary Priors
• Variable selection using Bayesian additive regression trees
• Bayesian estimation of sparse signals with a continuous spike-and-slab prior
• Bayesian approaches to variable selection in mixture models with application to disease clustering
• Prior distributions for objective Bayesian analysis
• Transport Monte Carlo: High-Accuracy Posterior Approximation via Random Transport
• Nearly optimal Bayesian shrinkage for high-dimensional regression
• Bayesian tensor regression
• The spike-and-slab LASSO
• Discussions
• Graph signal denoising using \(t\)-shrinkage priors
• Inference with normal-gamma prior distributions in regression problems
• Skinny Gibbs: a consistent and scalable Gibbs sampler for model selection
• Bayesian deconvolution and quantification of metabolites from \(J\)-resolved NMR spectroscopy
• Scalable approximate MCMC algorithms for the horseshoe prior
• Frequentist conditional variance for nonlinear mixed-effects models
• Robust sparse Bayesian infinite factor models
• Spatial Shrinkage Via the Product Independent Gaussian Process Prior
• Bayesian neural networks for selection of drug sensitive genes
• Revisiting Jeffreys’ Example: Bayes Test of the Normal Mean
• Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-differentiable Priors
• Fast and accurate variational inference for large Bayesian VARs with stochastic volatility
• Bayesian semiparametric multivariate density deconvolution
• Contraction properties of shrinkage priors in logistic regression
• Bayesian high-dimensional semi-parametric inference beyond sub-Gaussian errors
• Empirical estimates for heteroscedastic hierarchical dynamic normal models
• A hybrid scan Gibbs sampler for Bayesian models with latent variables
• Locally adaptive smoothing with Markov random fields and shrinkage priors
• Optimal Gaussian approximations to the posterior for log-linear models with Diaconis-Ylvisaker priors
• Bayesian Factor Analysis for Inference on Interactions

This page was built for publication: Dirichlet-Laplace priors for optimal shrinkage

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5367461)