Papers › Fast is better than free: Revisiting adversarial training

Fast is better than free: Revisiting adversarial training

12 Jan 2020ICLR 2020 1arXiv:2001.03994archive 2025-07-28

Eric Wong, Leslie Rice, J. Zico Kolter

Adversarial training, a method for learning robust deep networks, is typically assumed to be more expensive than traditional training due to the necessity of constructing adversarial examples via a first-order method like projected gradient decent (PGD). In this paper, we make the surprising discovery that it is possible to train empirically robust models using a much weaker and cheaper adversary, an approach that was previously believed to be ineffective, rendering the method no more costly than standard training in practice. Specifically, we show that adversarial training with the fast gradient sign method (FGSM), when combined with random initialization, is as effective as PGD-based training but has significantly lower cost. Furthermore we show that FGSM adversarial training can be further accelerated by using standard techniques for efficient training of deep networks, allowing us to learn a robust CIFAR10 classifier with 45% robust accuracy to PGD attacks with ϵ=8/255 in 6 minutes, and a robust ImageNet classifier with 43% robust accuracy at ϵ=2/255 in 12 hours, in comparison to past work based on "free" adversarial training which took 10 and 50 hours to reach the same respective thresholds. Finally, we identify a failure mode referred to as "catastrophic overfitting" which may have caused previous attempts to use FGSM adversarial training to fail. All code for reproducing the experiments in this paper as well as pretrained model weights are at https://github.com/locuslab/fast_adversarial.

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 32 code samples harvested from 7 repositories linked to this paper: 22 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 2 violated, 15 with no contract checked; 5 where Syntology's instrument failed) · 10 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: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 31 samples from 6 repositories, 22 ran (17 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 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 22 samples that ran has no recorded date.

locuslab/fast_adversarial officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
MetaSolver/icml2021 mentioned on GitHubpytorchBSD-3-ClauseSyntology: 11 samples, 7 ran (0 instrument-failed) · 4 unverified report
aliborji/Shapedefence mentioned on GitHubpytorchMITSyntology: 9 samples, 9 ran (1 instrument-failed) · 0 unverified report
aliborji/shapedefense mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
alirezaabdollahpour/CURE_fast_adversarial mentioned on GitHubpytorchSyntology: 5 samples, 3 ran (3 instrument-failed) · 2 unverified report
elijahcn/TextCNN-AdversarialTraining mentioned on GitHubpytorchMITSyntology: 3 samples, 1 ran (0 instrument-failed) · 2 unverified report
jokeryan/post_training mentioned on GitHubpytorchSyntology: not harvested report
juliagusak/neural-ode-metasolver mentioned on GitHubpytorchBSD-3-ClauseSyntology: not harvested report
rohban-lab/catastrophic_overfitting mentioned on GitHubpytorchSyntology: not harvested report
simon0987/Fast_FGSM mentioned on GitHubpytorchSyntology: not harvested report
tml-epfl/understanding-fast-adv-training mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 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

32 samples harvested: 22 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 2 violated, 15 with no contract checked; 5 where Syntology's instrument failed) · 10 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.

2ran · violated contract
5ran · our draft was wrong
15ran
10unverified

Licence: 9 of the 32 samples are pointer only, meaning Syntology does not serve that copy's text. For 1 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 7 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.

locuslab/fast_adversarial official repository 1 sample, 0 ran · 1 unverified
clamp in 78 papers
census: 7 implementations from 8 papers, 6 distinct outputs across 4 buckets, one shared input per bucket
MNIST/evaluate_mnist.py unverified no licence file found · pointer only · 8a93e041134b597a · report
MetaSolver/icml2021 community (archive-listed) 11 samples, 7 ran (0 instrument-failed) · 4 unverified
build_ButcherTableau_Heun sopa/src/solvers/rk_parametric_order2stage2.py ran BSD-3-Clause (permissive) · 5f605ede6476e2b2 · report fixture: signature dtype:dtype/0/?, device:device/0/? · returned tuple · level recorded 2026-09-24
build_ButcherTableau_Midpoint sopa/src/solvers/rk_parametric_order2stage2.py ran BSD-3-Clause (permissive) · ef2b42bfdaa3556e · report fixture: signature dtype:dtype/0/?, device:str/0/str · returned tuple · level recorded 2026-09-24
build_ButcherTableau_RK38 sopa/src/solvers/rk_parametric_order4stage4.py ran BSD-3-Clause (permissive) · 414bc3938ef8364c · report fixture: signature dtype:dtype/0/?, device:str/0/str · returned tuple · level recorded 2026-09-24
build_ButcherTableau_RKStandard sopa/src/solvers/rk_parametric_order4stage4.py ran BSD-3-Clause (permissive) · 2bc3009bc37610d0 · report fixture: signature dtype:dtype/0/?, device:device/0/? · returned tuple · level recorded 2026-09-24
clamp in 78 papers
census: 7 implementations from 8 papers, 6 distinct outputs across 4 buckets, one shared input per bucket
MegaAdversarial/src/attacks/fgsm.py ran fingerprinted BSD-3-Clause (permissive) · f1f502008abfb7b0 · report fixture: signature X:Tensor/2/float, lower_limit:float/0/float, upper_limit:float/0/float · returned Tensor · level recorded 2026-09-24
noise_params sopa/src/solvers/utils.py ran BSD-3-Clause (permissive) · 7d6882c87b7d7d18 · report fixture: signature mean_u:Tensor/0/float, mean_v:Tensor/0/float, std:float/0/float, bernoulli_p:float/0/float, noise_type:str/0/str, minimize_rk2_error:bool/0/bool · returned tuple · level recorded 2026-09-24
sample_noise in 10 papers sopa/src/solvers/utils.py ran BSD-3-Clause (permissive) · 1a22911f27fa7a50 · report fixture: signature mu:float/0/float, sigma:float/0/float, noise_type:str/0/str, size:int/0/int, device:str/0/str, minimize_rk2_error:bool/0/bool · returned Tensor · level recorded 2026-09-24
create_solver sopa/src/solvers/utils.py unverified BSD-3-Clause (permissive) · 412d129df08139d7 · report
get_cifar10_test_loader sopa/src/models/odenet_cifar10/data.py unverified BSD-3-Clause (permissive) · 842519d6c63098ed · report
get_cifar10_train_val_loaders sopa/src/models/odenet_cifar10/data.py unverified BSD-3-Clause (permissive) · 6d6817a15cfeccc0 · report
load_model in 479 papers sopa/src/models/utils.py unverified BSD-3-Clause (permissive) · 7ceaa0e5bd758ab3 · report
aliborji/Shapedefence community (archive-listed) 9 samples, 9 ran (1 instrument-failed) · 0 unverified
is_image_file in 157 papers make_imagenet_64_c.py ran · violated contract MIT (permissive) here; the stored copy is pointer only · 89c2cf3b97685684 · report fixture: signature filename:str/0/str · returned bool · level recorded, date not recorded
detect_edge_gtsrb edge_detector.py ran fingerprinted MIT (permissive) · 5f385a03a326c920 · report fixture: signature img:Tensor/3/float · returned ndarray · level recorded 2026-09-24
detect_edge_new edge_detector.py ran fingerprinted MIT (permissive) · daa0543c8e065130 · report fixture: signature img:Tensor/3/float · returned ndarray · level recorded 2026-09-24
detect_edge_new_cifar edge_detector.py ran fingerprinted MIT (permissive) · 9b843c30b475815c · report fixture: signature img:Tensor/3/float · returned ndarray · level recorded 2026-09-24
find_classes in 61 papers make_imagenet_64_c.py ran · our draft was wrong MIT (permissive) · 52d10a10fb72fb9f · report fixture: signature dir:str/0/str · returned tuple · level recorded 2026-09-24
get_norm_layer in 111 papers pix2pix-pytorch/networks.py ran MIT (permissive) · a116f92d34de08d0 · report fixture: signature norm_type:str/0/str · returned partial · level recorded 2026-09-24
get_scheduler in 122 papers pix2pix-pytorch/networks.py ran MIT (permissive) · 78865ef3023c58ad · report fixture: signature optimizer:Adam/0/?, opt:OptConfig/0/? · returned LambdaLR · level recorded 2026-09-24
init_net in 47 papers pix2pix-pytorch/networks.py ran MIT (permissive) · debf7fa6a54c7259 · report fixture: signature net:callable/0/?, init_type:str/0/str, init_gain:float/0/float, gpu_id:str/0/str · returned SimpleNet · level recorded 2026-09-24
make_dataset in 144 papers make_imagenet_64_c.py ran MIT (permissive) · 195faaace44890e8 · report fixture: signature dir:str/0/str, class_to_idx:dict/0/? · returned list · level recorded 2026-09-25
aliborji/shapedefense community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
auc in 21 papers test_noise_robustness.py ran · violated contract fingerprinted MIT (permissive) · 7a93b4bf9b9553b9 · report fixture: signature errs:list/1/float · returned float · level recorded 2026-09-22
alirezaabdollahpour/CURE_fast_adversarial community (archive-listed) 5 samples, 3 ran (3 instrument-failed) · 2 unverified
l0_mid_points attacks/fmn.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 89f1ad5572085e04 · report fixture: signature x0:Tensor/2/float, x1:Tensor/2/float, ε:Tensor/1/float · returned Tensor · level recorded 2026-09-22
l1_mid_points attacks/fmn.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 47bfdbebc768db6a · report fixture: signature x0:Tensor/2/float, x1:Tensor/2/float, ε:Tensor/1/float · returned Tensor · level recorded 2026-09-22
linf_projection_Alireza attacks/fmn.py ran · our draft was wrong fingerprinted no licence file found · pointer only · f6debb1c3f5dd7da · report fixture: signature δ:Tensor/3/float, ε:Tensor/1/float · returned Tensor · level recorded 2026-09-22
attack_pgd_Alireza train_CIFAR10.py unverified no licence file found · pointer only · b6a22c7b74d0b566 · report
zero_grad train_CIFAR10.py unverified no licence file found · pointer only · 5c6723ed7945b680 · report
elijahcn/TextCNN-AdversarialTraining community (archive-listed) 3 samples, 1 ran (0 instrument-failed) · 2 unverified
build_iterator utils.py ran MIT (permissive) · d2f33f1ecdf5b3bc · report fixture: signature dataset:list/1/tuple, config:Config/0/? · returned DatasetIterater · level recorded 2026-09-24
build_dataset in 151 papers utils.py unverified MIT (permissive) · c31db8461c9422cd · report
build_vocab in 36 papers utils.py unverified MIT (permissive) · 48d009a3bfb3919f · report
tml-epfl/understanding-fast-adv-training community (archive-listed) 2 samples, 1 ran (1 instrument-failed) · 1 unverified
PreActResNet18 in 38 papers models.py ran · our draft was wrong no licence file found · pointer only · fa636f649fc7a458 · report fixture: signature n_cls:int/0/int, cuda:bool/0/bool, half_prec:bool/0/bool · returned PreActResNet · level recorded 2026-09-22
get_model in 368 papers models.py unverified no licence file found · pointer only · cc55d19dea2beabe · report

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