Papers › Fast is better than free: Revisiting adversarial training
Fast is better than free: Revisiting adversarial training
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.
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):
get_harvested_code_for_paper(arxiv_id="2001.03994")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="2001.03994", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2001.03994")
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.
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.
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.
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
5f605ede6476e2b2 · report
fixture: signature dtype:dtype/0/?, device:device/0/? · returned tuple · level recorded 2026-09-24 ef2b42bfdaa3556e · report
fixture: signature dtype:dtype/0/?, device:str/0/str · returned tuple · level recorded 2026-09-24 414bc3938ef8364c · report
fixture: signature dtype:dtype/0/?, device:str/0/str · returned tuple · level recorded 2026-09-24 2bc3009bc37610d0 · report
fixture: signature dtype:dtype/0/?, device:device/0/? · returned tuple · level recorded 2026-09-24 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 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 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 412d129df08139d7 · report
842519d6c63098ed · report
6d6817a15cfeccc0 · report
7ceaa0e5bd758ab3 · report
89c2cf3b97685684 · report
fixture: signature filename:str/0/str · returned bool · level recorded, date not recorded 5f385a03a326c920 · report
fixture: signature img:Tensor/3/float · returned ndarray · level recorded 2026-09-24 daa0543c8e065130 · report
fixture: signature img:Tensor/3/float · returned ndarray · level recorded 2026-09-24 9b843c30b475815c · report
fixture: signature img:Tensor/3/float · returned ndarray · level recorded 2026-09-24 52d10a10fb72fb9f · report
fixture: signature dir:str/0/str · returned tuple · level recorded 2026-09-24 a116f92d34de08d0 · report
fixture: signature norm_type:str/0/str · returned partial · level recorded 2026-09-24 78865ef3023c58ad · report
fixture: signature optimizer:Adam/0/?, opt:OptConfig/0/? · returned LambdaLR · level recorded 2026-09-24 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 195faaace44890e8 · report
fixture: signature dir:str/0/str, class_to_idx:dict/0/? · returned list · level recorded 2026-09-25 7a93b4bf9b9553b9 · report
fixture: signature errs:list/1/float · returned float · level recorded 2026-09-22 89f1ad5572085e04 · report
fixture: signature x0:Tensor/2/float, x1:Tensor/2/float, ε:Tensor/1/float · returned Tensor · level recorded 2026-09-22 47bfdbebc768db6a · report
fixture: signature x0:Tensor/2/float, x1:Tensor/2/float, ε:Tensor/1/float · returned Tensor · level recorded 2026-09-22 f6debb1c3f5dd7da · report
fixture: signature δ:Tensor/3/float, ε:Tensor/1/float · returned Tensor · level recorded 2026-09-22 b6a22c7b74d0b566 · report
5c6723ed7945b680 · report
d2f33f1ecdf5b3bc · report
fixture: signature dataset:list/1/tuple, config:Config/0/? · returned DatasetIterater · level recorded 2026-09-24 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 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