Papers › Recurrent Independent Mechanisms

Recurrent Independent Mechanisms

24 Sep 2019ICLR 2021 1arXiv:1909.10893archive 2025-07-28

Anirudh Goyal, Alex Lamb, Jordan Hoffmann, Shagun Sodhani, Sergey Levine, Yoshua Bengio, Bernhard Schölkopf

Learning modular structures which reflect the dynamics of the environment can lead to better generalization and robustness to changes which only affect a few of the underlying causes. We propose Recurrent Independent Mechanisms (RIMs), a new recurrent architecture in which multiple groups of recurrent cells operate with nearly independent transition dynamics, communicate only sparingly through the bottleneck of attention, and are only updated at time steps where they are most relevant. We show that this leads to specialization amongst the RIMs, which in turn allows for dramatically improved generalization on tasks where some factors of variation differ systematically between training and evaluation.

PaperPDFConference PDFCodeCode 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 20 code samples: 19 from 3 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 20: 14 ran (of which 9 constructed an object rather than computing a result; 14 with no instrument failure: 1 honoured, 0 violated, 13 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: community (archive-listed): 19 samples from 3 repositories, 14 ran (14 with no instrument failure; 0 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. 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.

anirudh9119/RIMs mentioned on GitHubpytorchSyntology: 12 samples, 8 ran (0 instrument-failed) · 4 unverified report
dido1998/Recurrent-Independent-Mechanisms mentioned on GitHubpytorchSyntology: 4 samples, 3 ran (0 instrument-failed) · 1 unverified report
fuyuan-li/tensorflow-RIMs mentioned on GitHubtfSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 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

20 samples harvested: 14 ran (of which 9 constructed an object rather than computing a result; 14 with no instrument failure: 1 honoured, 0 violated, 13 with no contract checked; 0 where Syntology's instrument failed) · 6 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
13ran
6unverified

Licence: 20 of the 20 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. 1 sample is listed last, in a group without a repository. “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 1 sample 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.

anirudh9119/RIMs community (archive-listed) 12 samples, 8 ran (0 instrument-failed) · 4 unverified
BlockGRU blocks_atari/a2c_ppo_acktr/blocks_core.py ran no licence file found · pointer only · 779d2155a1554713 · report fixture: returned BlockGRU · level recorded 2026-09-22
BlockLSTM blocks_atari/a2c_ppo_acktr/blocks_core.py ran no licence file found · pointer only · c4c16932e4104a16 · report fixture: returned BlockLSTM · level recorded 2026-09-22
GroupLinearLayer blocks_atari/a2c_ppo_acktr/blocks_core.py ran no licence file found · pointer only · 80e31c08f502d673 · report fixture: returned GroupLinearLayer · level recorded 2026-09-22
ScaledDotProductAttention in 46 papers blocks_atari/a2c_ppo_acktr/blocks_core.py ran no licence file found · pointer only · cacc4a65e64179c2 · report fixture: returned ScaledDotProductAttention · level recorded 2026-09-22
Sparse_attention blocks_atari/a2c_ppo_acktr/blocks_core.py ran fingerprinted no licence file found · pointer only · cc1a9597719c4ced · report fixture: returned Sparse_attention · level recorded 2026-09-22
Sparse_grad_attention blocks_atari/a2c_ppo_acktr/blocks_core.py ran no licence file found · pointer only · af4c47bd3c2432c3 · report fixture: returned Sparse_grad_attention · level recorded 2026-09-22
blocked_grad blocks_atari/a2c_ppo_acktr/blocks_core.py ran no licence file found · pointer only · 534a1d4c71e7cd54 · report fixture: returned blocked_grad · level recorded 2026-09-22
repackage_hidden in 31 papers
census: 5 implementations from 7 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
event_based/train_copying.py ran · honoured contract fingerprinted no licence file found · pointer only · a81c5fb0b9535995 · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
BlocksCore blocks_atari/a2c_ppo_acktr/blocks_core.py unverified no licence file found · pointer only · 74ea1996d4a9ce45 · report
repackage_hidden in 31 papers
census: 5 implementations from 7 papers, 1 distinct output on one shared input
event_based/train_adding.py unverified no licence file found · pointer only · 70d200be038af543 · report
zero_matrix_elements blocks_atari/a2c_ppo_acktr/blocks_core.py unverified no licence file found · pointer only · 5b4032d119530f92 · report
dido1998/Recurrent-Independent-Mechanisms community (archive-listed) 4 samples, 3 ran (0 instrument-failed) · 1 unverified
GroupGRUCell RIM.py ran licence not identified · pointer only · 530004f559bbe61b · report fixture: signature x:Tensor/3/float, hidden:Tensor/3/float · returned Tensor · level recorded 2026-09-22
GroupLSTMCell RIM.py ran · metamorphic tier: deterministic licence not identified · pointer only · 70827a6ca3aea643 · report fixture: signature x:Tensor/3/float, hid_state:tuple/1/Tensor · returned tuple · level recorded 2026-09-22
GroupLinearLayer RIM.py ran fingerprinted licence not identified · pointer only · 3dbce3a83b09b310 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
RIMCell RIM.py unverified licence not identified · pointer only · 406669b7f6420d33 · report
fuyuan-li/tensorflow-RIMs community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
GroupLSTMCell tfRIM.py ran licence not identified · pointer only · 03ed40c46c6366e9 · report fixture: returned GroupLSTMCell · level recorded 2026-09-22
GroupLinearLayer tfRIM.py ran licence not identified · pointer only · b1f19581fe3bab53 · report fixture: signature inputs:EagerTensor/3/float · returned EagerTensor · level recorded 2026-09-24
RIMCell tfRIM.py ran licence not identified · pointer only · b440016392f1a553 · report fixture: returned RIMCell · level recorded 2026-09-22
Repository not located 1 sample, 0 ran · 1 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
none_or_str in 23 papers repository not located unverified licence of this copy not recorded · 2fc6fdb85f13dc41 · 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. 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.

3 assemblies on this paper: 3 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

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

harvested Assembly for Recurrent Independent Mechanisms (graph method: Recurrent Independent Mechanisms). 10 bricks; individually on a fixture: 7 ran (of which 7 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified; every one of the 7 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 3 siblings

  1. BlocksCore entry · level 0 · unverified · 74ea1996d4a9ce45
  2. BlockGRU helper · level 1 · ran · 779d2155a1554713
  3. BlockLSTM helper · level 1 · ran · c4c16932e4104a16
  4. MultiHeadAttention helper · level 0 · unverified · 2aaad11d08c7dcef
  5. blocked_grad helper · level 1 · ran · 534a1d4c71e7cd54
  6. zero_matrix_elements helper · level 0 · unverified · 5b4032d119530f92
  7. GroupLinearLayer helper · level 1 · ran · 80e31c08f502d673
  8. ScaledDotProductAttention helper · level 1 · ran · cacc4a65e64179c2
  9. Sparse_attention helper · level 1 · ran · cc1a9597719c4ced
  10. Sparse_grad_attention helper · level 1 · ran · af4c47bd3c2432c3

harvested Assembly for Recurrent Independent Mechanisms (graph method: Recurrent Independent Mechanisms). 3 bricks; individually on a fixture: 3 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. RIMCell entry · level 1 · ran · b440016392f1a553
  2. GroupLSTMCell helper · level 1 · ran · 03ed40c46c6366e9
  3. GroupLinearLayer helper · level 1 · ran · b1f19581fe3bab53

harvested Assembly for Recurrent Independent Mechanisms (graph method: Recurrent Independent Mechanisms). 5 bricks; individually on a fixture: 4 ran (of which 1 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. RIMCell entry · level 0 · unverified · 406669b7f6420d33
  2. GroupGRUCell helper · level 1 · ran · 530004f559bbe61b
  3. GroupLSTMCell helper · level 1 · ran · 70827a6ca3aea643
  4. GroupLinearLayer helper · level 1 · ran · 3dbce3a83b09b310
  5. blocked_grad helper · level 1 · ran · 534a1d4c71e7cd54

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="1909.10893") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Atari Games

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Atari Games Atari 2600 Beam Rider RIMs-PPO Score 5320 #39 of 49 Archive leaderboard report
Atari Games Atari 2600 Up and Down RIMs-PPO Score 390000 #10 of 44 Archive leaderboard report
Atari Games Atari 2600 Zaxxon RIMs-PPO Score 15000 #16 of 41 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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