Rényi's $α$-divergence variational Bayes for spike-and-slab high-dimensional linear regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bsila, Chadi, Tang, Yiqi, Wang, Kaiwen, Heyer, Laurie
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911294742855680
author Bsila, Chadi
Tang, Yiqi
Wang, Kaiwen
Heyer, Laurie
author_facet Bsila, Chadi
Tang, Yiqi
Wang, Kaiwen
Heyer, Laurie
contents Sparse high-dimensional linear regression is a central problem in statistics, where the goal is often variable selection and/or coefficient estimation. We propose a mean-field variational Bayes approximation for sparse regression with spike-and-slab Laplace priors that replaces the standard Kullback-Leibler (KL) divergence objective with the Rényi's $α$ divergence: a one-parameter generalization of KL divergence indexed by $α\in (0, \infty) \setminus \{1\}$ that allows flexibility between zero-forcing and mass-covering behavior. We derive coordinate ascent variational inference (CAVI) updates via a second-order delta method and develop a stochastic variational inference algorithm based on a Monte Carlo surrogate Rényi lower bound. In simulations, our two methods perform comparably to state-of-the-art Bayesian variable selection procedures across a range of sparsity configurations and $α$ values for both variable selection and estimation, and our numerical results illustrate how different choices of $α$ can be advantageous in different sparsity configurations.
format Preprint
id arxiv_https___arxiv_org_abs_2512_00627
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rényi's $α$-divergence variational Bayes for spike-and-slab high-dimensional linear regression
Bsila, Chadi
Tang, Yiqi
Wang, Kaiwen
Heyer, Laurie
Methodology
Computation
Sparse high-dimensional linear regression is a central problem in statistics, where the goal is often variable selection and/or coefficient estimation. We propose a mean-field variational Bayes approximation for sparse regression with spike-and-slab Laplace priors that replaces the standard Kullback-Leibler (KL) divergence objective with the Rényi's $α$ divergence: a one-parameter generalization of KL divergence indexed by $α\in (0, \infty) \setminus \{1\}$ that allows flexibility between zero-forcing and mass-covering behavior. We derive coordinate ascent variational inference (CAVI) updates via a second-order delta method and develop a stochastic variational inference algorithm based on a Monte Carlo surrogate Rényi lower bound. In simulations, our two methods perform comparably to state-of-the-art Bayesian variable selection procedures across a range of sparsity configurations and $α$ values for both variable selection and estimation, and our numerical results illustrate how different choices of $α$ can be advantageous in different sparsity configurations.
title Rényi's $α$-divergence variational Bayes for spike-and-slab high-dimensional linear regression
topic Methodology
Computation
url https://arxiv.org/abs/2512.00627