Stein Variational Gradient Descent dynamics for highly concentrated kernels

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Carrillo, José A., Skrzeczkowski, Jakub, Warnett, Jethro
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913090702934016
author Carrillo, José A.
Skrzeczkowski, Jakub
Warnett, Jethro
author_facet Carrillo, José A.
Skrzeczkowski, Jakub
Warnett, Jethro
contents Stein Variational Gradient Descent (SVGD) is a widely used in practice algorithm for scalable sampling with deterministic particle updates. We study its behavior in the singular limit where the kernel bandwidth tends to zero. In this regime, we show that the nonlocal SVGD dynamics converge to a local evolution equation that can be formally interpreted as a Wasserstein gradient flow with quadratic mobility. We analyze this singular limit in two settings: integrable kernels and weighted kernels. In the weighted case, the proof is supported by recently established Stein-log-Sobolev inequalities, which provide the necessary functional control. Overall, our results clarify how SVGD collapses from a nonlocal interacting particle system to a local gradient-flow dynamics as the kernel concentrates.
format Preprint
id arxiv_https___arxiv_org_abs_2605_03627
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Stein Variational Gradient Descent dynamics for highly concentrated kernels
Carrillo, José A.
Skrzeczkowski, Jakub
Warnett, Jethro
Analysis of PDEs
35Q62, 35Q68, 35B40, 62-08, 62D05
Stein Variational Gradient Descent (SVGD) is a widely used in practice algorithm for scalable sampling with deterministic particle updates. We study its behavior in the singular limit where the kernel bandwidth tends to zero. In this regime, we show that the nonlocal SVGD dynamics converge to a local evolution equation that can be formally interpreted as a Wasserstein gradient flow with quadratic mobility. We analyze this singular limit in two settings: integrable kernels and weighted kernels. In the weighted case, the proof is supported by recently established Stein-log-Sobolev inequalities, which provide the necessary functional control. Overall, our results clarify how SVGD collapses from a nonlocal interacting particle system to a local gradient-flow dynamics as the kernel concentrates.
title Stein Variational Gradient Descent dynamics for highly concentrated kernels
topic Analysis of PDEs
35Q62, 35Q68, 35B40, 62-08, 62D05
url https://arxiv.org/abs/2605.03627