Diffusion Models for Reinforcement Learning: A Survey
It provides an overview for researchers in RL and generative modeling, but is incremental as it synthesizes existing work rather than presenting new findings.
This survey examines how diffusion models, which offer superior sample quality and training stability, are being applied to address challenges in reinforcement learning (RL) and improve RL solutions.
Diffusion models surpass previous generative models in sample quality and training stability. Recent works have shown the advantages of diffusion models in improving reinforcement learning (RL) solutions. This survey aims to provide an overview of this emerging field and hopes to inspire new avenues of research. First, we examine several challenges encountered by RL algorithms. Then, we present a taxonomy of existing methods based on the roles of diffusion models in RL and explore how the preceding challenges are addressed. We further outline successful applications of diffusion models in various RL-related tasks. Finally, we conclude the survey and offer insights into future research directions. We are actively maintaining a GitHub repository for papers and other related resources in utilizing diffusion models in RL: https://github.com/apexrl/Diff4RLSurvey.