Saved in:
Bibliographic Details
Main Authors: Braun, Cornelius V., Lange, Robert T., Toussaint, Marc
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.10390
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911507193790464
author Braun, Cornelius V.
Lange, Robert T.
Toussaint, Marc
author_facet Braun, Cornelius V.
Lange, Robert T.
Toussaint, Marc
contents Stein Variational Gradient Descent (SVGD) is a highly efficient method to sample from an unnormalized probability distribution. However, the SVGD update relies on gradients of the log-density, which may not always be available. Existing gradient-free versions of SVGD make use of simple Monte Carlo approximations or gradients from surrogate distributions, both with limitations. To improve gradient-free Stein variational inference, we combine SVGD steps with evolution strategy (ES) updates. Our results demonstrate that the resulting algorithm generates high-quality samples from unnormalized target densities without requiring gradient information. Compared to prior gradient-free SVGD methods, we find that the integration of the ES update in SVGD significantly improves the performance on multiple challenging benchmark problems.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10390
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Stein Variational Evolution Strategies
Braun, Cornelius V.
Lange, Robert T.
Toussaint, Marc
Machine Learning
Artificial Intelligence
Neural and Evolutionary Computing
Stein Variational Gradient Descent (SVGD) is a highly efficient method to sample from an unnormalized probability distribution. However, the SVGD update relies on gradients of the log-density, which may not always be available. Existing gradient-free versions of SVGD make use of simple Monte Carlo approximations or gradients from surrogate distributions, both with limitations. To improve gradient-free Stein variational inference, we combine SVGD steps with evolution strategy (ES) updates. Our results demonstrate that the resulting algorithm generates high-quality samples from unnormalized target densities without requiring gradient information. Compared to prior gradient-free SVGD methods, we find that the integration of the ES update in SVGD significantly improves the performance on multiple challenging benchmark problems.
title Stein Variational Evolution Strategies
topic Machine Learning
Artificial Intelligence
Neural and Evolutionary Computing
url https://arxiv.org/abs/2410.10390