Papers › On the interaction between supervision and self-play in emergent communication

On the interaction between supervision and self-play in emergent communication

4 Feb 2020ICLR 2020 1arXiv:2002.01093archive 2025-07-28

Ryan Lowe, Abhinav Gupta, Jakob Foerster, Douwe Kiela, Joelle Pineau

A promising approach for teaching artificial agents to use natural language involves using human-in-the-loop training. However, recent work suggests that current machine learning methods are too data inefficient to be trained in this way from scratch. In this paper, we investigate the relationship between two categories of learning signals with the ultimate goal of improving sample efficiency: imitating human language data via supervised learning, and maximizing reward in a simulated multi-agent environment via self-play (as done in emergent communication), and introduce the term supervised self-play (S2P) for algorithms using both of these signals. We find that first training agents via supervised learning on human data followed by self-play outperforms the converse, suggesting that it is not beneficial to emerge languages from scratch. We then empirically investigate various S2P schedules that begin with supervised learning in two environments: a Lewis signaling game with symbolic inputs, and an image-based referential game with natural language descriptions. Lastly, we introduce population based approaches to S2P, which further improves the performance over single-agent methods.

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 16 code samples harvested from 1 repository linked to this paper: 11 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 7 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 3 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 11 ran (4 with no instrument failure; 7 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 on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

backpropper/s2p officialmentioned in paperpytorchSyntology: 16 samples, 11 ran (7 instrument-failed) · 5 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

16 samples harvested: 11 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 7 where Syntology's instrument failed) · 5 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
4ran · our draft was wrong
3ran; the contract check could not run
3ran
5unverified

Licence: 0 of the 16 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 backpropper/s2p. “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.

backpropper/s2p official repository 16 samples, 11 ran (7 instrument-failed) · 5 unverified
Beholder ibr_game/finetune.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 7a6f9cb1a39ca0b3 · report fixture: returned Beholder · level recorded 2026-09-22
Listener ibr_game/finetune.py ran · metamorphic tier: deterministic MIT (permissive) · 9ecb51a47b201507 · report fixture: returned Listener · level recorded 2026-09-22
Speaker ibr_game/finetune.py ran · metamorphic tier: deterministic MIT (permissive) · 9f509297c82bb3a7 · report fixture: returned Speaker · level recorded 2026-09-22
filter_caps ibr_game/finetune.py ran · our draft was wrong MIT (permissive) · 02b107d78fd0fca7 · report fixture: signature captions:list/1/list, images:list/1/Tensor, w2i:dict/0/?, perc:float/0/float · returned tuple · level recorded 2026-09-22
get_s2p_batch ibr_game/finetune.py ran; the contract check could not run (RAISES) MIT (permissive) · 5a0aaf20a82cf665 · report fixture: signature seed_train_images:Tensor/3/float, seed_train_captions:list/1/ndarray, args:Args/0/?, batch_size:NoneType/0/? · returned tuple · level recorded 2026-09-22
gumbel_softmax in 82 papers
census: 19 implementations from 20 papers, 12 distinct outputs across 2 buckets, one shared input per bucket
ibr_game/finetune.py ran; the contract check could not run (RAISES) · metamorphic tier: deterministic MIT (permissive) · 779be7e9ddb772f0 · report fixture: signature logits:Tensor/2/float, temp:float/0/float, hard:bool/0/bool · returned tuple · level recorded 2026-09-22
gumbel_softmax_sample in 49 papers
census: 14 implementations from 16 papers, 4 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
ibr_game/finetune.py ran · honoured contract fingerprinted MIT (permissive) · 626e470a849ef8d9 · report fixture: signature logits:Tensor/2/float, temp:float/0/float · returned Tensor · level recorded 2026-09-22
sample_gumbel in 62 papers ibr_game/finetune.py ran · our draft was wrong MIT (permissive) · eb35bf807af0e213 · report fixture: signature shape:tuple/1/int, eps:float/0/float · returned Tensor · level recorded 2026-09-22
trim_caps ibr_game/finetune.py ran · our draft was wrong MIT (permissive) · 7b8ff1df3ed9fbd2 · report fixture: signature caps:list/1/list, minlen:int/0/int, maxlen:int/0/int · returned list · level recorded 2026-09-22
truncate_captions ibr_game/finetune.py ran; the contract check could not run (RAISES) · metamorphic tier: deterministic MIT (permissive) · bfdb7c5db11d3441 · report fixture: signature train_captions:list/1/list, valid_captions:list/1/list, test_captions:list/1/list, w2i:dict/0/?, i2w:dict/0/? · returned tuple · level recorded 2026-09-22
truncate_dicts ibr_game/finetune.py ran · our draft was wrong MIT (permissive) · d004ca3c465884a6 · report fixture: signature w2i:dict/0/?, i2w:dict/0/?, trunc_size:int/0/int · returned tuple · level recorded 2026-09-22
SpeakerListener ibr_game/finetune.py unverified MIT (permissive) · 057dbfd6c2a4131d · report
load_listener ibr_game/finetune.py unverified MIT (permissive) · 6509889b8433dd2e · report
main in 235 papers ibr_game/finetune.py unverified MIT (permissive) · 5dffc621e220f31f · report
save_to_file ibr_game/finetune.py unverified MIT (permissive) · 975549298f10836c · report
torch_save_to_file ibr_game/finetune.py unverified MIT (permissive) · 409eb80fda6b01b0 · 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 Supervised Self-Play (S2P) (graph method: Supervised Self-Play (S2P)). 16 bricks; individually on a fixture: 11 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 7 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.

  1. main entry · level 0 · unverified · 5dffc621e220f31f
  2. SpeakerListener model · level 0 · unverified · 057dbfd6c2a4131d
  3. filter_caps helper · level 1 · ran · our draft was wrong · 02b107d78fd0fca7
  4. get_s2p_batch helper · level 1 · ran; the contract check could not run (RAISES) · 5a0aaf20a82cf665
  5. load_listener helper · level 0 · unverified · 6509889b8433dd2e
  6. save_to_file helper · level 0 · unverified · 975549298f10836c
  7. torch_save_to_file helper · level 0 · unverified · 409eb80fda6b01b0
  8. trim_caps helper · level 1 · ran · our draft was wrong · 7b8ff1df3ed9fbd2
  9. truncate_captions helper · level 1 · ran; the contract check could not run (RAISES) · bfdb7c5db11d3441
  10. truncate_dicts helper · level 1 · ran · our draft was wrong · d004ca3c465884a6
  11. Beholder model · level 1 · ran · 7a6f9cb1a39ca0b3
  12. Listener model · level 1 · ran · 9ecb51a47b201507
  13. Speaker helper · level 1 · ran · 9f509297c82bb3a7
  14. gumbel_softmax helper · level 1 · ran; the contract check could not run (RAISES) · 779be7e9ddb772f0
  15. gumbel_softmax_sample helper · level 1 · ran · honoured contract · 626e470a849ef8d9
  16. sample_gumbel helper · level 1 · ran · our draft was wrong · eb35bf807af0e213

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="2002.01093") returns these assemblies with each brick's role and wiring (how to connect).

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