Papers › Maximum-Entropy Adversarial Data Augmentation for Improved Generalization and Robustness

Maximum-Entropy Adversarial Data Augmentation for Improved Generalization and Robustness

15 Oct 2020NeurIPS 2020 12arXiv:2010.08001archive 2025-07-28

Long Zhao, Ting Liu, Xi Peng, Dimitris Metaxas

Adversarial data augmentation has shown promise for training robust deep neural networks against unforeseen data shifts or corruptions. However, it is difficult to define heuristics to generate effective fictitious target distributions containing "hard" adversarial perturbations that are largely different from the source distribution. In this paper, we propose a novel and effective regularization term for adversarial data augmentation. We theoretically derive it from the information bottleneck principle, which results in a maximum-entropy formulation. Intuitively, this regularization term encourages perturbing the underlying source distribution to enlarge predictive uncertainty of the current model, so that the generated "hard" adversarial perturbations can improve the model robustness during training. Experimental results on three standard benchmarks demonstrate that our method consistently outperforms the existing state of the art by a statistically significant margin.

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 22 code samples harvested from 1 repository linked to this paper: 17 ran (of which 11 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) · 5 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 22 samples from 1 repository, 17 ran (12 with no instrument failure; 5 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 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.

garyzhao/ME-ADA officialmentioned in papermentioned on GitHubpytorchSyntology: 22 samples, 17 ran (5 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

22 samples harvested: 17 ran (of which 11 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) · 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.

5ran · our draft was wrong
12ran
5unverified

Licence: 0 of the 22 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 garyzhao/ME-ADA. “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 6 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.

garyzhao/ME-ADA official repository 22 samples, 17 ran (5 instrument-failed) · 5 unverified
AllConvNet model_cifar.py ran BSD-3-Clause (permissive) · 2190125be2989110 · report fixture: returned AllConvNet · level recorded 2026-09-22
BasicBlock in 126 papers
census: 6 implementations from 6 papers, 5 distinct outputs on one shared input
model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · e28e6b33589030ee · report fixture: returned BasicBlock · level recorded 2026-09-22
Bottleneck in 98 papers model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · b7b0351f60ac3a4a · report fixture: returned Bottleneck · level recorded 2026-09-22
CifarResNeXt model_cifar.py ran BSD-3-Clause (permissive) · 9a69a7117b29ff25 · report fixture: returned CifarResNeXt · level recorded 2026-09-22
Denormalise model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · 4a47aedd3d544b45 · report fixture: returned Denormalise · level recorded 2026-09-22
DenseNet in 5 papers model_cifar.py ran BSD-3-Clause (permissive) · 1e6effa8af67f1f1 · report fixture: returned DenseNet · level recorded 2026-09-22
GELU in 21 papers model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · a3653639a4779e6c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-24
NetworkBlock in 6 papers model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · 6aa4a39a16fbc4fc · report fixture: returned NetworkBlock · level recorded 2026-09-22
ResNeXtBottleneck model_cifar.py ran BSD-3-Clause (permissive) · d1f712f2674c7e26 · report fixture: returned ResNeXtBottleneck · level recorded 2026-09-22
SingleLayer model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · 24cce59d906bf43b · report fixture: returned SingleLayer · level recorded 2026-09-22
Transition in 7 papers model_cifar.py ran fingerprinted BSD-3-Clause (permissive) · 8710e0de3e2bf007 · report fixture: returned Transition · level recorded 2026-09-22
WideResNet in 13 papers model_cifar.py ran BSD-3-Clause (permissive) · 64abc4d84960db53 · report fixture: returned WideResNet · level recorded 2026-09-22
densenet model_cifar.py ran · our draft was wrong BSD-3-Clause (permissive) · 9e21fa66980b9f96 · report fixture: signature growth_rate:int/0/int, depth:int/0/int, num_classes:int/0/int · returned DenseNet · level recorded 2026-09-22
entropy_loss in 42 papers
census: 5 implementations from 5 papers, 2 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
model_cifar.py ran · our draft was wrong fingerprinted BSD-3-Clause (permissive) · 10487302bd997fba · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
make_layers in 155 papers model_cifar.py ran · our draft was wrong BSD-3-Clause (permissive) · fc16595d232916da · report fixture: signature cfg:list/1/int · returned Sequential · level recorded 2026-09-22
resnext29 model_cifar.py ran · our draft was wrong BSD-3-Clause (permissive) · 9fbec066275ba93d · report fixture: signature num_classes:int/0/int, cardinality:int/0/int, base_width:int/0/int · returned CifarResNeXt · level recorded 2026-09-22
sgd in 8 papers model_cifar.py ran · our draft was wrong BSD-3-Clause (permissive) · e54d44b5a4d7e951 · report fixture: signature parameters:generator/0/?, lr:float/0/float, weight_decay:float/0/float, momentum:float/0/float · returned SGD · level recorded 2026-09-22
ModelADA model_cifar.py unverified BSD-3-Clause (permissive) · bdde82053558f603 · report
ModelBaseline model_cifar.py unverified BSD-3-Clause (permissive) · b91d260808e5ccec · report
ModelMEADA model_cifar.py unverified BSD-3-Clause (permissive) · c29a896aa9b7f6c6 · report
fix_all_seed model_cifar.py unverified BSD-3-Clause (permissive) · 43b4e7134e89de82 · report
write_log in 6 papers model_cifar.py unverified BSD-3-Clause (permissive) · f71d878c2fdfce45 · 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 Maximum-Entropy Adversarial Data Augmentation (graph method: Maximum-Entropy Adversarial Data Augmentation). 22 bricks; individually on a fixture: 17 ran (of which 11 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) · 5 unverified. The assembly has not been run end-to-end.

  1. ModelMEADA entry · level 0 · unverified · c29a896aa9b7f6c6
  2. ModelADA helper · level 0 · unverified · bdde82053558f603
  3. entropy_loss loss · level 1 · ran · our draft was wrong · 10487302bd997fba
  4. sgd helper · level 1 · ran · our draft was wrong · e54d44b5a4d7e951
  5. write_log helper · level 0 · unverified · f71d878c2fdfce45
  6. Denormalise helper · level 1 · ran · 4a47aedd3d544b45
  7. ModelBaseline helper · level 0 · unverified · b91d260808e5ccec
  8. AllConvNet helper · level 1 · ran · 2190125be2989110
  9. WideResNet helper · level 1 · ran · 64abc4d84960db53
  10. densenet helper · level 1 · ran · our draft was wrong · 9e21fa66980b9f96
  11. fix_all_seed helper · level 0 · unverified · 43b4e7134e89de82
  12. resnext29 helper · level 1 · ran · our draft was wrong · 9fbec066275ba93d
  13. make_layers helper · level 1 · ran · our draft was wrong · fc16595d232916da
  14. BasicBlock helper · level 1 · ran · e28e6b33589030ee
  15. NetworkBlock helper · level 1 · ran · 6aa4a39a16fbc4fc
  16. DenseNet helper · level 1 · ran · 1e6effa8af67f1f1
  17. CifarResNeXt helper · level 1 · ran · 9a69a7117b29ff25
  18. ResNeXtBottleneck helper · level 1 · ran · d1f712f2674c7e26
  19. GELU helper · level 1 · ran · a3653639a4779e6c
  20. Bottleneck helper · level 1 · ran · b7b0351f60ac3a4a
  21. SingleLayer helper · level 1 · ran · 24cce59d906bf43b
  22. Transition helper · level 1 · ran · 8710e0de3e2bf007

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="2010.08001") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Data Augmentation

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