Papers › Towards Deep Learning Models Resistant to Adversarial Attacks

Towards Deep Learning Models Resistant to Adversarial Attacks

19 Jun 2017ICLR 2018 1arXiv:1706.06083archive 2025-07-28

Aleksander Madry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, Adrian Vladu

Recent work has demonstrated that deep neural networks are vulnerable to adversarial examples---inputs that are almost indistinguishable from natural data and yet classified incorrectly by the network. In fact, some of the latest findings suggest that the existence of adversarial attacks may be an inherent weakness of deep learning models. To address this problem, we study the adversarial robustness of neural networks through the lens of robust optimization. This approach provides us with a broad and unifying view on much of the prior work on this topic. Its principled nature also enables us to identify methods for both training and attacking neural networks that are reliable and, in a certain sense, universal. In particular, they specify a concrete security guarantee that would protect against any adversary. These methods let us train networks with significantly improved resistance to a wide range of adversarial attacks. They also suggest the notion of security against a first-order adversary as a natural and broad security guarantee. We believe that robustness against such well-defined classes of adversaries is an important stepping stone towards fully resistant deep learning models. Code and pre-trained models are available at https://github.com/MadryLab/mnist_challenge and https://github.com/MadryLab/cifar10_challenge.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1706.06083")

Code

Syntology Ran 9 of 17 code samples harvested from 3 repositories linked to this paper; 8 have no recorded run. Of those that ran: 2 ran · honoured contract; 6 ran · our draft was wrong; 1 ran · fixture could not drive it.

By repository: community (archive-listed): 9 samples from 3 repositories, 3 ran; 8 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.

59 repositories listed; official and paper-mentioned ones first.

MadryLab/cifar10_challenge officialmentioned in papermentioned on GitHubtfMIT report
MadryLab/mnist_challenge officialmentioned in papermentioned on GitHubtfMIT report
EPFL-VILAB/XDEnsembles mentioned on GitHubpytorch report
Hadisalman/robust-verify-benchmark mentioned on GitHubpytorch report
Jeffkang-94/pytorch-adversarial-attack mentioned on GitHubpytorch report
KnowledgeDiscovery/FaceSec mentioned on GitHubpytorch report
P2333/Max-Mahalanobis-Training mentioned on GitHubtfApache-2.0 report
SafiyaJan/Attacking-Neural-Networks mentioned on GitHubpytorch report
VishaalMK/VectorDefense mentioned on GitHubtf report
Zoky-2020/Set-level_Guidance_Attack mentioned on GitHubpytorchMIT report
abahram77/mnistChallenge mentioned on GitHubtf report
abahram77/mnist_challenge mentioned on GitHubtf report
amerch/CIFAR100-Training mentioned on GitHubpytorch report
arobey1/advbench mentioned on GitHubpytorch report
arobey1/mbrdl mentioned on GitHubpytorchNOASSERTION report
bethgelab/cifar10_challenge mentioned on GitHubtf report
bingcheng45/hnr-extension mentioned on GitHubtf report
boyellow/adaad mentioned on GitHubpytorchMIT report
cdluminate/advrank mentioned on GitHubpytorch report
cdluminate/advrank-pub mentioned on GitHubpytorch report
cleverhans-lab/cleverhans mentioned on GitHubtfMIT report
cs-giung/course-dl-TP mentioned on GitHubpytorch report
eldadp100/cnn_course_final mentioned on GitHubpytorch report
henry8527/GCE mentioned on GitHubpytorch report
hope-yao/robust_attention_cifar mentioned on GitHubtf report
hrdwsong/TDLMR2AA-Paddle mentioned on GitHubpaddle report
jokeryan/post_training mentioned on GitHubpytorch report
khieu/cifar10_challenge mentioned on GitHubtf report
locuslab/convex_adversarial mentioned on GitHubpytorch report
locuslab/robust_overfitting mentioned on GitHubpytorch report
loes5307/vocaladversary2022 mentioned on GitHubpytorch report
luizgh/adversarial_signatures mentioned on GitHubpytorch report
matanbt/attack-tabular mentioned on GitHub report
microsoft/distance-learner mentioned on GitHubpytorch report
openai/cleverhans mentioned on GitHubtf report
peck94/cann-detector mentioned on GitHubtf report
revbucket/mister_ed mentioned on GitHubpytorch report
scenarri/s2m-tea mentioned on GitHubpytorchApache-2.0 report
tensorflow/cleverhans mentioned on GitHubtfMIT report
thomashopkins32/PGDAdversarialLearning mentioned on GitHubpytorch report
ucsb-nlp-chang/textgrad mentioned on GitHubpytorch report
val-iisc/flss mentioned on GitHubpytorch report
zibojia/rslad mentioned on GitHubpytorchMIT report
zjfheart/Friendly-Adversarial-Training mentioned on GitHubpytorch 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

17 samples harvested; 9 ran; 2 honoured the contract we drafted; 8 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
6ran · our draft was wrong
1ran · fixture could not drive it
8unverified

Licence: 13 of the 17 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; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

attack_pgd zibojia/rslad/rslad_loss.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 57555b28180e36f9 · report
kl_loss zibojia/rslad/mobilenet_v2_rslad_cifar10.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · 9960f247d801fa2a · report
rslad_inner_loss zibojia/rslad/rslad_loss.py community (archive-listed) ran · our draft was wrong MIT (permissive) · d4a9a69d1dba912f · report
MART_loss zjfheart/Friendly-Adversarial-Training/FAT_for_MART.py community (archive-listed) unverified no licence file found · pointer only · f0d4b8c8f55455d1 · report
TRADES_loss zjfheart/Friendly-Adversarial-Training/FAT_for_TRADES.py community (archive-listed) unverified no licence file found · pointer only · 310d3d6ec0329c05 · report
adjust_tau zjfheart/Friendly-Adversarial-Training/FAT.py community (archive-listed) unverified no licence file found · pointer only · ed8707d32cea6f61 · report
adjust_tau zjfheart/Friendly-Adversarial-Training/FAT_for_TRADES.py community (archive-listed) unverified no licence file found · pointer only · 9257b5b1f6195ecf · report
adjust_tau zjfheart/Friendly-Adversarial-Training/FAT_for_MART.py community (archive-listed) unverified no licence file found · pointer only · 102fe6c0dc1d18bc · report
evaluate_baseline locuslab/convex_adversarial/examples/trainer.py community (archive-listed) unverified MIT (permissive) · 3918fb83399cf118 · report
conv3x3 identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · 583f9780bdd00a45 · report
mixup_criterion identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · 33ba52fc17e89516 · report
mixup_data identical code first harvested elsewhere ran · fixture could not drive it licence of this copy not recorded · b20f1357b1d8dbf8 · report
mixup_data identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · 11b0ec76b88d8553 · report
pseudorandom_target identical code first harvested elsewhere ran · honoured contract fingerprinted licence of this copy not recorded · dc3545f86c25dbe2 · report
pseudorandom_target_image identical code first harvested elsewhere ran · honoured contract fingerprinted licence of this copy not recorded · 108e0e470de94dfe · report
clamp identical code first harvested elsewhere unverified licence of this copy not recorded · 8a93e041134b597a · report
get_image identical code first harvested elsewhere unverified licence of this copy not recorded · 197a1c1869bf9116 · report

Tasks

Adversarial AttackAdversarial DefenseAdversarial RobustnessDeep LearningImage ClassificationObject DetectionPart-Of-Speech TaggingRobust classificationSound Event DetectionVideo Quality Assessment

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Adversarial Attack CIFAR-10 AdvTraining [madry2018] Attack: PGD20 48.440 #2 of 6 Archive leaderboard report
Part-Of-Speech Tagging Morphosyntactic-analysis-dataset MyBert BLEX 77.21 #1 of 1 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