PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 13, 20240 citationsOpen Access

DiffPoGAN: Diffusion Policies with Generative Adversarial Networks for Offline Reinforcement Learning

View Full Paper
XHXuemin HuLSLi ShenYXYingfen Xu

Key Points

  • DiffPoGAN yields superior policy performance while addressing the extrapolation error issue prevalent in offline reinforcement learning.
  • In experiments, DiffPoGAN consistently outperformed existing state-of-the-art methods, enhancing action distribution coverage significantly.
  • This assessment utilized datasets from the D4RL benchmark, focusing on offline reinforcement learning scenarios requiring robust policy exploration techniques.  — The findings underscore the necessity for better exploration constraints in RL models.

Abstract

Offline reinforcement learning (RL) can learn optimal policies from pre-collected offline datasets without interacting with the environment, but the sampled actions of the agent cannot often cover the action distribution under a given state, resulting in the extrapolation error issue. Recent works address this issue by employing generative adversarial networks (GANs). However, these methods often suffer from insufficient constraints on policy exploration and inaccurate representation of behavior policies. Moreover, the generator in GANs fails in fooling the discriminator while maximizing the expected returns of a policy. Inspired by the diffusion, a generative model with powerful feature expressiveness, we propose a new offline RL method named Diffusion Policies with Generative Adversarial Networks (DiffPoGAN). In this approach, the diffusion serves as the policy generator to generate diverse distributions of actions, and a regularization method based on maximum likelihood estimation (MLE) is developed to generate data that approximate the distribution of behavior policies. Besides, we introduce an additional regularization term based on the discriminator output to effectively constrain policy exploration for policy improvement. Comprehensive experiments are conducted on the datasets for deep data-driven reinforcement learning (D4RL), and experimental results show that DiffPoGAN outperforms state-of-the-art methods in offline RL.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Hu et al. (2024) studied this question.

synapsesocial.com/papers/68e64f88b6db6435875e00ddhttps://doi.org/10.48550/arxiv.2406.09089
Ask AI
Helpful
Bookmark
Share
View Full Paper