Papers › Implicit competitive regularization in GANs

Implicit competitive regularization in GANs

13 Oct 2019ICML 2020 1arXiv:1910.05852archive 2025-07-28

Florian Schäfer, Hongkai Zheng, Anima Anandkumar

To improve the stability of GAN training we need to understand why they can produce realistic samples. Presently, this is attributed to properties of the divergence obtained under an optimal discriminator. This argument has a fundamental flaw: If we do not impose regularity of the discriminator, it can exploit visually imperceptible errors of the generator to always achieve the maximal generator loss. In practice, gradient penalties are used to regularize the discriminator. However, this needs a metric on the space of images that captures visual similarity. Such a metric is not known, which explains the limited success of gradient penalties in stabilizing GANs. We argue that the performance of GANs is instead due to the implicit competitive regularization (ICR) arising from the simultaneous optimization of generator and discriminator. ICR promotes solutions that look real to the discriminator and thus leverages its inductive biases to generate realistic images. We show that opponent-aware modelling of generator and discriminator, as present in competitive gradient descent (CGD), can significantly strengthen ICR and thus stabilize GAN training without explicit regularization. In our experiments, we use an existing implementation of WGAN-GP and show that by training it with CGD we can improve the inception score (IS) on CIFAR10 for a wide range of scenarios, without any hyperparameter tuning. The highest IS is obtained by combining CGD with the WGAN-loss, without any explicit regularization.

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 14 code samples harvested from 2 repositories linked to this paper: 13 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; 3 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 14 samples from 2 repositories, 13 ran (10 with no instrument failure; 3 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 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

devzhk/Implicit-Competitive-Regularization mentioned on GitHubpytorchApache-2.0Syntology: 9 samples, 8 ran (3 instrument-failed) · 1 unverified report
devzhk/cgds-package mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Competitive Gradient Descent (arXiv:1905.12103; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
wagenaartje/torch-cgd mentioned on GitHubpytorchMITSyntology: 5 samples, 5 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.

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

14 samples harvested: 13 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; 3 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
3ran · our draft was wrong
9ran
1unverified

Licence: 2 of the 14 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 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”. “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.

devzhk/Implicit-Competitive-Regularization community (archive-listed) 9 samples, 8 ran (3 instrument-failed) · 1 unverified
all_gather in 119 papers distributed.py ran Apache-2.0 (permissive) · ed03ba54397adbe4 · report fixture: signature data:dict/0/? · returned list · level recorded 2026-09-26
d_logistic_loss in 15 papers
census: 2 implementations from 7 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 1, kind float, dtype float32 · arg 2: rank 1, kind float, dtype float32
losses.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 3651db4e5ae8d189 · report fixture: signature real_pred:Tensor/1/float, fake_pred:Tensor/1/float · returned Tensor · level recorded 2026-09-22
d_r1_loss in 10 papers losses.py ran Apache-2.0 (permissive) · 0143fef228211c97 · report fixture: signature real_pred:Tensor/2/float, real_img:Tensor/4/float · returned Tensor · level recorded 2026-09-24
get_loss in 107 papers
census: 5 implementations from 5 papers, 5 distinct outputs across 4 buckets, one shared input per bucket
losses.py ran Apache-2.0 (permissive) · b3537003a80eee42 · report fixture: signature name:str/0/str, g_loss:bool/0/bool, d_real:Tensor/2/float, d_fake:Tensor/2/float, gp_weight:int/0/int, l2_weight:float/0/float, D:callable/0/? · returned Tensor · level recorded 2026-09-24
gram_schmidt in 32 papers GANs/layers.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · c41836f3e51fa8f1 · report fixture: signature x:Tensor/2/float, ys:list/1/Tensor · returned Tensor · level recorded 2026-09-22
power_iteration in 30 papers GANs/layers.py ran · our draft was wrong Apache-2.0 (permissive) · 28d3a35157f2916d · report fixture: signature W:Tensor/2/float, u_:list/1/Tensor, update:bool/0/bool, eps:float/0/float · returned tuple · level recorded 2026-09-25
reduce_loss_dict in 18 papers distributed.py ran Apache-2.0 (permissive) · 5144ad456049f8ce · report fixture: signature loss_dict:dict/0/? · returned dict · level recorded 2026-09-26
reduce_sum in 34 papers
census: 4 implementations from 5 papers, 4 distinct outputs across 4 buckets, one shared input per bucket
distributed.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 3f0a1105ef251d78 · report fixture: signature tensor:Tensor/2/float · returned Tensor · level recorded 2026-09-24
proj in 38 papers GANs/layers.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 4015ab1f3ab9b881 · report
wagenaartje/torch-cgd community (archive-listed) 5 samples, 5 ran (0 instrument-failed) · 0 unverified
apply_given_rotation torch_cgd/solvers/gmres.py ran MIT (permissive) · b7914ea8af7be96d · report fixture: signature H:Tensor/2/float, cs:list/1/Tensor, ss:list/1/Tensor, j:int/0/int · returned tuple · level recorded 2026-09-24
arnoldi torch_cgd/solvers/gmres.py ran MIT (permissive) · 5c0dbc25ffbe6494 · report fixture: signature vec:Tensor/1/float, V:Tensor/2/float, H:Tensor/2/float, j:int/0/int · returned Tensor · level recorded 2026-09-24
cal_rotation torch_cgd/solvers/gmres.py ran MIT (permissive) · 9ccb04835431ae7d · report fixture: signature a:Tensor/0/float, b:Tensor/0/float · returned tuple · level recorded 2026-09-24
ith_element torch_cgd/utils.py ran MIT (permissive) · 302d868d4be5c1c5 · report fixture: signature params:list/1/Tensor, index:int/0/int · returned Tensor · level recorded 2026-09-24
vectorize in 12 papers torch_cgd/utils.py ran MIT (permissive) · 480f12ea8540663b · report fixture: signature l:list/1/Tensor · returned Tensor · level recorded 2026-09-24

Tasks

Image Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Convolution

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