Saved in:
Bibliographic Details
Main Authors: Zhang, Yuyang, Hu, Yang, Dai, Bo, Li, Na
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.23870
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917175163355136
author Zhang, Yuyang
Hu, Yang
Dai, Bo
Li, Na
author_facet Zhang, Yuyang
Hu, Yang
Dai, Bo
Li, Na
contents Soft actor-critic (SAC) is a popular algorithm for max-entropy reinforcement learning. In practice, the energy-based policies in SAC are often approximated using simple policy classes for efficiency, sacrificing the expressiveness and robustness. In this paper, we propose a variant of the SAC algorithm that parameterizes the policy with flow-based models, leveraging their rich expressiveness. In the algorithm, we evaluate the flow-based policy utilizing the instantaneous change-of-variable technique and update the policy with an online variant of flow matching developed in this paper. This online variant, termed importance sampling flow matching (ISFM), enables policy update with only samples from a user-specified sampling distribution rather than the unknown target distribution. We develop a theoretical analysis of ISFM, characterizing how different choices of sampling distributions affect the learning efficiency. Finally, we conduct a case study of our algorithm on the max-entropy linear quadratic regulator problems, demonstrating that the proposed algorithm learns the optimal action distribution.
format Preprint
id arxiv_https___arxiv_org_abs_2512_23870
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Max-Entropy Reinforcement Learning with Flow Matching and A Case Study on LQR
Zhang, Yuyang
Hu, Yang
Dai, Bo
Li, Na
Machine Learning
Soft actor-critic (SAC) is a popular algorithm for max-entropy reinforcement learning. In practice, the energy-based policies in SAC are often approximated using simple policy classes for efficiency, sacrificing the expressiveness and robustness. In this paper, we propose a variant of the SAC algorithm that parameterizes the policy with flow-based models, leveraging their rich expressiveness. In the algorithm, we evaluate the flow-based policy utilizing the instantaneous change-of-variable technique and update the policy with an online variant of flow matching developed in this paper. This online variant, termed importance sampling flow matching (ISFM), enables policy update with only samples from a user-specified sampling distribution rather than the unknown target distribution. We develop a theoretical analysis of ISFM, characterizing how different choices of sampling distributions affect the learning efficiency. Finally, we conduct a case study of our algorithm on the max-entropy linear quadratic regulator problems, demonstrating that the proposed algorithm learns the optimal action distribution.
title Max-Entropy Reinforcement Learning with Flow Matching and A Case Study on LQR
topic Machine Learning
url https://arxiv.org/abs/2512.23870