Papers › Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning

Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning

12 Aug 2022arXiv:2208.06193archive 2025-07-28

Zhendong Wang, Jonathan J Hunt, Mingyuan Zhou

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.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Syntology holds a reference implementation of Diffusion Qlearning (Diffusion-QL), generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). For this implementation, level 3 is recorded, but its report lists no property test that passed on both. No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='Diffusion Qlearning (Diffusion-QL)') (how to connect); the free check have('2208.06193') needs no token.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 18 code samples harvested from 3 repositories linked to this paper: 11 ran (of which 5 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 3 samples from 1 repository, 1 ran (1 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 15 samples from 2 repositories, 10 ran (10 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

zhendong-wang/diffusion-policies-for-offline-rl officialmentioned in papermentioned on GitHubpytorchSyntology: 3 samples, 1 ran (0 instrument-failed) · 2 unverified report
quantumiracle/consistency_model_for_reinforcement_learning mentioned on GitHubpytorchSyntology: 5 samples, 2 ran (0 instrument-failed) · 3 unverified report
zzmtsvv/rl_task mentioned on GitHubpytorchSyntology: 10 samples, 8 ran (0 instrument-failed) · 2 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

18 samples harvested: 11 ran (of which 5 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
10ran
7unverified

Licence: 10 of the 18 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 2 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

zhendong-wang/diffusion-policies-for-offline-rl official repository 3 samples, 1 ran (0 instrument-failed) · 2 unverified
EMA in 18 papers agents/ql_diffusion.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · cb039993bd16f94c · report fixture: signature ma_model:callable/0/?, current_model:callable/0/? · returned NoneType · level recorded 2026-09-22
Diffusion in 21 papers agents/ql_diffusion.py unverified Apache-2.0 (permissive) · a8e86dce13aa99a1 · report
Diffusion_QL agents/ql_diffusion.py unverified Apache-2.0 (permissive) · 56d3224268de9651 · report
quantumiracle/consistency_model_for_reinforcement_learning community (archive-listed) 5 samples, 2 ran (0 instrument-failed) · 3 unverified
EMA in 18 papers agents/ql_diffusion.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 19732228c46fa1ba · report fixture: signature ma_model:callable/0/?, current_model:callable/0/? · returned NoneType · level recorded 2026-09-22
MLP in 375 papers agents/ql_diffusion.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 6a06af32e9295cda · report fixture: signature x:Tensor/2/float, time:Tensor/2/float, state:Tensor/2/float · returned Tensor · level recorded 2026-09-22
Diffusion in 21 papers agents/ql_diffusion.py unverified Apache-2.0 (permissive) · 49c674376f6d739f · report
Diffusion_QL agents/ql_diffusion.py unverified Apache-2.0 (permissive) · f3e8db467ebcdc65 · report
WeightedHuber agents/ql_diffusion.py unverified Apache-2.0 (permissive) · a193aa42d45269f9 · report
zzmtsvv/rl_task community (archive-listed) 10 samples, 8 ran (0 instrument-failed) · 2 unverified
DDPM in 15 papers diffusion_ql/dql.py ran fingerprinted no licence file found · pointer only · 08e51c2d2c41c81f · report fixture: returned DDPM · level recorded 2026-09-22
EnsembledCritic in 7 papers diffusion_ql/dql.py ran no licence file found · pointer only · d6fcbdadf8cb4e3f · report fixture: returned EnsembledCritic · level recorded 2026-09-22
EnsembledLinear in 5 papers diffusion_ql/dql.py ran no licence file found · pointer only · ac150cae89f0e687 · report fixture: returned EnsembledLinear · level recorded 2026-09-22
MLP in 375 papers diffusion_ql/dql.py ran fingerprinted no licence file found · pointer only · 911033f1c5373d15 · report fixture: signature action:Tensor/2/float, state:Tensor/2/float, time:Tensor/1/float · returned Tensor · level recorded 2026-09-24
PositionalEncoding in 126 papers
census: 13 implementations from 9 papers, 9 distinct outputs across 5 buckets, one shared input per bucket
diffusion_ql/dql.py ran no licence file found · pointer only · ef90a18431b489f9 · report fixture: returned PositionalEncoding · level recorded 2026-09-22
WeightedL2 in 8 papers diffusion_ql/dql.py ran fingerprinted no licence file found · pointer only · 480e0c2c2219727e · report fixture: signature prediction:Tensor/2/float, target:Tensor/2/float, weights:Tensor/2/float · returned Tensor · level recorded 2026-09-24
dql_config diffusion_ql/dql.py ran no licence file found · pointer only · 9cbe616246577b15 · report fixture: returned dql_config · level recorded 2026-09-22
vp_beta_schedule in 10 papers diffusion_ql/dql.py ran · honoured contract no licence file found · pointer only · 25a711883c5a59fe · report fixture: signature timesteps:int/0/int, dtype:dtype/0/? · returned Tensor · level recorded 2026-09-22
DiffusionQL diffusion_ql/dql.py unverified no licence file found · pointer only · 8136ea50760183ad · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another. Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

4 assemblies on this paper: 3 harvested, 1 generated, listed in 2 groups (siblings lifted in one batch form one group).

generated Assembly for Diffusion Qlearning (Diffusion-QL) (graph method: Diffusion Qlearning (Diffusion-QL)), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer; level 3 is recorded, but its report lists no property test that passed on both.

label matches by construction: the code was generated for this method (not judged by a model)

  1. polyak_average_update entry · level 3 · ran · honoured contract · 21e738aa59fdd95c2ece78a0411be6a997984f01518494d5c71ab169bbcc260f

3 sibling assemblies for Diffusion Qlearning (Diffusion-QL) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Diffusion Qlearning (Diffusion-QL) (graph method: Diffusion Qlearning (Diffusion-QL)). 10 bricks; individually on a fixture: 8 ran (of which 5 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. DiffusionQL entry · level 0 · unverified · 8136ea50760183ad
  2. DDPM model · level 1 · ran · 08e51c2d2c41c81f
  3. EnsembledCritic model · level 1 · ran · d6fcbdadf8cb4e3f
  4. MLP model · level 1 · ran · 911033f1c5373d15
  5. dql_config helper · level 1 · ran · 9cbe616246577b15
  6. WeightedL2 helper · level 1 · ran · 480e0c2c2219727e
  7. extract helper · level 0 · unverified · ba56ef20ff6743ce
  8. vp_beta_schedule scheduler · level 1 · ran · honoured contract · 25a711883c5a59fe
  9. EnsembledLinear helper · level 1 · ran · ac150cae89f0e687
  10. PositionalEncoding helper · level 1 · ran · ef90a18431b489f9

harvested Assembly for Diffusion Qlearning (Diffusion-QL) (graph method: Diffusion Qlearning (Diffusion-QL)). 15 bricks; individually on a fixture: 11 ran (of which 4 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. Diffusion_QL entry · level 0 · unverified · 56d3224268de9651
  2. Critic model · level 1 · ran · 5bdc9227d5668ad4
  3. Diffusion helper · level 0 · unverified · a8e86dce13aa99a1
  4. EMA helper · level 1 · ran · cb039993bd16f94c
  5. MLP helper · level 1 · ran · cb1c89fe9b73e635
  6. Progress helper · level 1 · ran · 33d1a23a0c89bc5b
  7. Silent helper · level 1 · ran · 98e65aa869e5b6c6
  8. cosine_beta_schedule helper · level 1 · ran · honoured contract · b6113e0f43155a33
  9. extract helper · level 0 · unverified · 09c8479d9a5b3e06
  10. linear_beta_schedule helper · level 1 · ran · honoured contract · 63d21c7afce1a9e5
  11. vp_beta_schedule helper · level 1 · ran · honoured contract · 47c7ce0ace6e7c47
  12. SinusoidalPosEmb helper · level 1 · ran · 0bd2edd74d66fbd9
  13. WeightedLoss helper · level 0 · unverified · 728e142be8c63b8b
  14. WeightedL1 helper · level 1 · ran · 20545bf1089ab44e
  15. WeightedL2 helper · level 1 · ran · 3b2b8b659d67b0af

harvested Assembly for Diffusion Qlearning (Diffusion-QL) (graph method: Diffusion Qlearning (Diffusion-QL)). 16 bricks; individually on a fixture: 11 ran (of which 3 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. Diffusion_QL entry · level 0 · unverified · f3e8db467ebcdc65
  2. Critic model · level 1 · ran · 5bdc9227d5668ad4
  3. Diffusion model · level 0 · unverified · 49c674376f6d739f
  4. EMA helper · level 1 · ran · 19732228c46fa1ba
  5. MLP helper · level 1 · ran · 6a06af32e9295cda
  6. Progress helper · level 1 · ran · 33d1a23a0c89bc5b
  7. Silent helper · level 1 · ran · 98e65aa869e5b6c6
  8. cosine_beta_schedule helper · level 1 · ran · honoured contract · b6113e0f43155a33
  9. extract helper · level 0 · unverified · 09c8479d9a5b3e06
  10. linear_beta_schedule helper · level 1 · ran · honoured contract · 63d21c7afce1a9e5
  11. vp_beta_schedule helper · level 1 · ran · honoured contract · 47c7ce0ace6e7c47
  12. SinusoidalPosEmb helper · level 1 · ran · 0bd2edd74d66fbd9
  13. WeightedLoss helper · level 0 · unverified · 0749c0edc53a9387
  14. WeightedHuber helper · level 0 · unverified · a193aa42d45269f9
  15. WeightedL1 helper · level 1 · ran · a195e5fbb1ffc8f2
  16. WeightedL2 helper · level 1 · ran · 04bd445ebe6af984

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2208.06193") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

D4RLOffline RLQ-LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

DiffusionQ-Learning

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections