Papers › Stabilizing Transformers for Reinforcement Learning

Stabilizing Transformers for Reinforcement Learning

13 Oct 2019ICML 2020 1arXiv:1910.06764archive 2025-07-28

Emilio Parisotto, H. Francis Song, Jack W. Rae, Razvan Pascanu, Caglar Gulcehre, Siddhant M. Jayakumar, Max Jaderberg, Raphael Lopez Kaufman, Aidan Clark, Seb Noury, Matthew M. Botvinick, Nicolas Heess, Raia Hadsell

Owing to their ability to both effectively integrate information over long time horizons and scale to massive amounts of data, self-attention architectures have recently shown breakthrough success in natural language processing (NLP), achieving state-of-the-art results in domains such as language modeling and machine translation. Harnessing the transformer's ability to process long time horizons of information could provide a similar performance boost in partially observable reinforcement learning (RL) domains, but the large-scale transformers used in NLP have yet to be successfully applied to the RL setting. In this work we demonstrate that the standard transformer architecture is difficult to optimize, which was previously observed in the supervised learning setting but becomes especially pronounced with RL objectives. We propose architectural modifications that substantially improve the stability and learning speed of the original Transformer and XL variant. The proposed architecture, the Gated Transformer-XL (GTrXL), surpasses LSTMs on challenging memory environments and achieves state-of-the-art results on the multi-task DMLab-30 benchmark suite, exceeding the performance of an external memory architecture. We show that the GTrXL, trained using the same losses, has stability and performance that consistently matches or exceeds a competitive LSTM baseline, including on more reactive tasks where memory is less critical. GTrXL offers an easy-to-train, simple-to-implement but substantially more expressive architectural alternative to the standard multi-layer LSTM ubiquitously used for RL agents in partially observable environments.

PaperPDFConference PDFCodeCode Syntology ran

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 3 code samples harvested from 1 repository linked to this paper: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 3 samples from 1 repository, 3 ran (2 with no instrument failure; 1 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 2 of its repositories that Syntology harvested for other papers:

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

JunnYu/x-transformers-paddle mentioned on GitHubjaxApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Swin Transformer V2: Scaling Up Capacity and Resolution (arXiv:2111.09883; that paper's own run record: 17 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 3 violated, 11 with no contract checked; 3 where Syntology's instrument failed) · 13 unverified); nothing checks that they implement this paper's method. report
jerrodparker20/adaptive-transformers-in-rl mentioned on GitHubpytorchSyntology: not harvested report
lucidrains/compressive-transformer-pytorch mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (1 instrument-failed) · 0 unverified report
dhruvramani/Transformers-RL pytorchMITSyntology: not harvested report
opendilab/DI-engine pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Hindsight Experience Replay (arXiv:1707.01495; that paper's own run record: 16 ran (of which 10 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 5 where Syntology's instrument failed) · 0 unverified), Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context (arXiv:1901.02860; that paper's own run record: 67 ran (of which 37 constructed an object rather than computing a result; 49 with no instrument failure: 4 honoured, 1 violated, 44 with no contract checked; 18 where Syntology's instrument failed) · 76 unverified), Decision Transformer: Reinforcement Learning via Sequence Modeling (arXiv:2106.01345; that paper's own run record: 17 ran (of which 10 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 5 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); 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

3 samples harvested: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 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.

2ran · violated contract
1ran · our draft was wrong

Licence: 3 of the 3 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 lucidrains/compressive-transformer-pytorch. 3 samples whose link to this paper records no repository are listed under one of this paper's own repositories, because identical code was harvested from it for another 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”. “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.

lucidrains/compressive-transformer-pytorch community (archive-listed) 3 samples, 3 ran (1 instrument-failed) · 0 unverified
cast_tuple in 58 papers compressive_transformer_pytorch/compressive_transformer_pytorch.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · violated contract fingerprinted licence of this copy not recorded · 369eced62e96d690 · report fixture: signature el:int/0/int · returned tuple · level recorded 2026-09-22
default in 430 papers compressive_transformer_pytorch/compressive_transformer_pytorch.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · violated contract licence of this copy not recorded · a1bda7590dd9a4d2 · report fixture: signature x:NoneType/0/?, val:callable/0/? · returned str · level recorded 2026-09-22
to in 15 papers compressive_transformer_pytorch/compressive_transformer_pytorch.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · our draft was wrong fingerprinted licence of this copy not recorded · 7417c98099ec4e15 · report fixture: signature t:Tensor/2/float · returned dict · level recorded 2026-09-22

Tasks

General Reinforcement LearningLanguage ModelingLanguage ModellingMachine TranslationPartially Observable Reinforcement LearningReinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: GTrXL

Absolute Position EncodingsAdamAdaptive Input RepresentationsAdaptive SoftmaxAttentionBPECosine AnnealingDense ConnectionsDropoutGRUGTrXLLSTMLabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSPEEDSigmoid ActivationSoftmaxTanh ActivationTransformerTransformer-XLVariational Dropout

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