Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:2208.06193

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Machine Learning

arXiv:2208.06193 (cs)
[Submitted on 12 Aug 2022 (v1), last revised 25 Aug 2023 (this version, v3)]

Title:Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning

Authors:Zhendong Wang, Jonathan J Hunt, Mingyuan Zhou
View a PDF of the paper titled Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning, by Zhendong Wang and 2 other authors
View PDF
Abstract:Offline reinforcement learning (RL), which aims to learn an optimal policy using a previously collected static dataset, is an important paradigm of RL. Standard RL methods often perform poorly in this regime due to the function approximation errors on out-of-distribution actions. While a variety of regularization methods have been proposed to mitigate this issue, they are often constrained by policy classes with limited expressiveness that can lead to highly suboptimal solutions. In this paper, we propose representing the policy as a diffusion model, a recent class of highly-expressive deep generative models. We introduce Diffusion Q-learning (Diffusion-QL) that utilizes a conditional diffusion model to represent the policy. In our approach, we learn an action-value function and we add a term maximizing action-values into the training loss of the conditional diffusion model, which results in a loss that seeks optimal actions that are near the behavior policy. We show the expressiveness of the diffusion model-based policy, and the coupling of the behavior cloning and policy improvement under the diffusion model both contribute to the outstanding performance of Diffusion-QL. We illustrate the superiority of our method compared to prior works in a simple 2D bandit example with a multimodal behavior policy. We then show that our method can achieve state-of-the-art performance on the majority of the D4RL benchmark tasks.
Comments: ICLR 2023
Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
Cite as: arXiv:2208.06193 [cs.LG]
  (or arXiv:2208.06193v3 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2208.06193
arXiv-issued DOI via DataCite

Submission history

From: Mingyuan Zhou [view email]
[v1] Fri, 12 Aug 2022 09:54:11 UTC (876 KB)
[v2] Mon, 3 Oct 2022 20:57:30 UTC (700 KB)
[v3] Fri, 25 Aug 2023 19:39:32 UTC (803 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning, by Zhendong Wang and 2 other authors
  • View PDF
  • TeX Source
  • Other Formats
view license
Current browse context:
cs.LG
< prev   |   next >
new | recent | 2022-08
Change to browse by:
cs
stat
stat.ML

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
IArxiv Recommender (What is IArxiv?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack