Papers › Zero-Shot Reinforcement Learning from Low Quality Data

Zero-Shot Reinforcement Learning from Low Quality Data

26 Sep 2023arXiv:2309.15178archive 2025-07-28

Scott Jeen, Tom Bewley, Jonathan M. Cullen

Zero-shot reinforcement learning (RL) promises to provide agents that can perform any task in an environment after an offline, reward-free pre-training phase. Methods leveraging successor measures and successor features have shown strong performance in this setting, but require access to large heterogenous datasets for pre-training which cannot be expected for most real problems. Here, we explore how the performance of zero-shot RL methods degrades when trained on small homogeneous datasets, and propose fixes inspired by conservatism, a well-established feature of performant single-task offline RL algorithms. We evaluate our proposals across various datasets, domains and tasks, and show that conservative zero-shot RL algorithms outperform their non-conservative counterparts on low quality datasets, and perform no worse on high quality datasets. Somewhat surprisingly, our proposals also outperform baselines that get to see the task during training. Our code is available via https://enjeeneer.io/projects/zero-shot-rl/ .

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.

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

Code

Syntology 18 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 0 where Syntology's instrument failed) · 8 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 18 samples from 1 repository, 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

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.

enjeeneer/conservative-world-models officialmentioned in papermentioned on GitHubpytorchSyntology: 18 samples, 10 ran (0 instrument-failed) · 8 unverified report
enjeeneer/zero-shot-rl officialmentioned in papermentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Towards Robust Zero-Shot Reinforcement Learning (arXiv:2510.15382; that paper's own run record: 13 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 5 honoured, 0 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 9 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

18 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 0 where Syntology's instrument failed) · 8 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
9ran
8unverified

Licence: 0 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 enjeeneer/conservative-world-models. “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”. “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.

enjeeneer/conservative-world-models official repository 18 samples, 10 ran (0 instrument-failed) · 8 unverified
AbstractActor agents/cfb/agent.py ran MIT (permissive) · 676367224566988e · report fixture: signature observation:Tensor/2/float, std:float/0/float · returned TruncatedNormal · level recorded 2026-09-24
AbstractGaussianActor agents/cfb/agent.py ran MIT (permissive) · 350547f0b8b706cb · report fixture: signature observation:Tensor/2/float · returned SquashedNormal · level recorded 2026-09-24
AbstractPreprocessor agents/cfb/agent.py ran MIT (permissive) · a9f70659526f31de · report fixture: signature concatenation:Tensor/2/float · returned Tensor · level recorded 2026-09-24
BackwardModel agents/cfb/agent.py ran MIT (permissive) · 815b8b7187def028 · report fixture: signature observation:Tensor/2/float · returned Tensor · level recorded 2026-09-24
BackwardRepresentation agents/cfb/agent.py ran fingerprinted MIT (permissive) · ff9f5d55817b5b9e · report fixture: signature observation:Tensor/2/float · returned Tensor · level recorded 2026-09-24
ForwardModel in 6 papers agents/cfb/agent.py ran · metamorphic tier: invariant MIT (permissive) · 23d435a38e160c8d · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
SquashedNormal agents/cfb/agent.py ran MIT (permissive) · 6360b567a0112530 · report fixture: signature sample_shape:Size/1/int · returned Tensor · level recorded 2026-09-24
TanhTransform in 5 papers agents/cfb/agent.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 18d2cc67d612e1e4 · report fixture: signature x:Tensor/2/float, y:Tensor/2/float · returned Tensor · level recorded 2026-09-22
TruncatedNormal in 7 papers agents/cfb/agent.py ran · metamorphic tier: well formed MIT (permissive) · eb6b6a9c095395c9 · report fixture: signature clip:float/0/float, sample_shape:Size/1/int · returned Tensor · level recorded 2026-09-22
schedule in 32 papers agents/cfb/agent.py ran · honoured contract fingerprinted MIT (permissive) · 53f6373e323067d0 · report fixture: signature schdl:str/0/str, step:int/0/int · returned float64 · level recorded 2026-09-22
AbstractAgent in 5 papers agents/cfb/agent.py unverified MIT (permissive) · 5c45e7cf6e639135 · report
AbstractMLP agents/cfb/agent.py unverified MIT (permissive) · 0952e55bb0dcf25c · report
ActorModel agents/cfb/agent.py unverified MIT (permissive) · aa653345ad8ad268 · report
Batch in 5 papers agents/cfb/agent.py unverified MIT (permissive) · b940ca501e37511f · report
CFB agents/cfb/agent.py unverified MIT (permissive) · c28fc24e49523491 · report
FB agents/cfb/agent.py unverified MIT (permissive) · d2fc051560d214e9 · report
ForwardBackwardRepresentation agents/cfb/agent.py unverified MIT (permissive) · 64523b02a6612ce0 · report
ForwardRepresentation agents/cfb/agent.py unverified MIT (permissive) · 2b3a474296d0ab6a · 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. 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.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for conservative zero-shot RL algorithms (graph method: conservative zero-shot RL algorithms). 18 bricks; individually on a fixture: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 0 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. CFB entry · level 0 · unverified · c28fc24e49523491
  2. FB helper · level 0 · unverified · d2fc051560d214e9
  3. AbstractAgent helper · level 0 · unverified · 5c45e7cf6e639135
  4. AbstractGaussianActor helper · level 1 · ran · 350547f0b8b706cb
  5. ActorModel helper · level 0 · unverified · aa653345ad8ad268
  6. Batch helper · level 0 · unverified · b940ca501e37511f
  7. ForwardBackwardRepresentation model · level 0 · unverified · 64523b02a6612ce0
  8. schedule helper · level 1 · ran · honoured contract · 53f6373e323067d0
  9. AbstractMLP helper · level 0 · unverified · 0952e55bb0dcf25c
  10. SquashedNormal helper · level 1 · ran · 6360b567a0112530
  11. AbstractActor helper · level 1 · ran · 676367224566988e
  12. AbstractPreprocessor helper · level 1 · ran · a9f70659526f31de
  13. BackwardRepresentation helper · level 1 · ran · ff9f5d55817b5b9e
  14. ForwardRepresentation helper · level 0 · unverified · 2b3a474296d0ab6a
  15. TanhTransform helper · level 1 · ran · 18d2cc67d612e1e4
  16. TruncatedNormal helper · level 1 · ran · eb6b6a9c095395c9
  17. BackwardModel helper · level 1 · ran · 815b8b7187def028
  18. ForwardModel helper · level 1 · ran · 23d435a38e160c8d

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="2309.15178") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Offline RLReinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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