Papers › Imbalanced Gradients: A Subtle Cause of Overestimated Adversarial Robustness

Imbalanced Gradients: A Subtle Cause of Overestimated Adversarial Robustness

24 Jun 2020arXiv:2006.13726archive 2025-07-28

Xingjun Ma, Linxi Jiang, Hanxun Huang, Zejia Weng, James Bailey, Yu-Gang Jiang

Evaluating the robustness of a defense model is a challenging task in adversarial robustness research. Obfuscated gradients have previously been found to exist in many defense methods and cause a false signal of robustness. In this paper, we identify a more subtle situation called Imbalanced Gradients that can also cause overestimated adversarial robustness. The phenomenon of imbalanced gradients occurs when the gradient of one term of the margin loss dominates and pushes the attack towards to a suboptimal direction. To exploit imbalanced gradients, we formulate a Margin Decomposition (MD) attack that decomposes a margin loss into individual terms and then explores the attackability of these terms separately via a two-stage process. We also propose a multi-targeted and ensemble version of our MD attack. By investigating 24 defense models proposed since 2018, we find that 11 models are susceptible to a certain degree of imbalanced gradients and our MD attack can decrease their robustness evaluated by the best standalone baseline attack by more than 1%. We also provide an in-depth investigation on the likely causes of imbalanced gradients and effective countermeasures. Our code is available at https://github.com/HanxunH/MDAttack.

PaperPDFCodeCode 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="2006.13726")

Code

Syntology Ran 4 of 9 code samples harvested from 1 repository linked to this paper; 5 have no recorded run. Of those that ran: 1 ran · our draft was wrong; 3 ran with no contract checked.

By repository: official repository: 9 samples from 1 repository, 4 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

hanxunh/mdattack officialmentioned in papermentioned on GitHubpytorchMIT 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

9 samples harvested; 4 ran; 0 honoured the contract we drafted; 5 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.

1ran · our draft was wrong
3ran
5unverified

Licence: 0 of the 9 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 hanxunh/mdattack. “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.

margin_loss hanxunh/mdattack/attacks/PGD.py official repository ran · our draft was wrong MIT (permissive) · 20c10282e0a610a7 · report
projection_l1 hanxunh/mdattack/attacks/fab_projections.py official repository ran fingerprinted MIT (permissive) · 9647e1400ee744e4 · report
projection_l2 hanxunh/mdattack/attacks/fab_projections.py official repository ran fingerprinted MIT (permissive) · a345677b413ef5ca · report
projection_linf hanxunh/mdattack/attacks/fab_projections.py official repository ran fingerprinted MIT (permissive) · b8bdb6e496b4b5fa · report
adv_check_and_update hanxunh/mdattack/attacks/utils.py official repository unverified MIT (permissive) · 4dcc85b6c29ae57b · report
cw_loss hanxunh/mdattack/attacks/PGD.py official repository unverified MIT (permissive) · 74ec6769bff8da35 · report
load_json hanxunh/mdattack/collect_results.py official repository unverified MIT (permissive) · 44930808e0a51ad4 · report
one_hot_tensor hanxunh/mdattack/attacks/utils.py official repository unverified MIT (permissive) · f65d4de4b1f07c14 · report
setup_logger hanxunh/mdattack/util.py official repository unverified MIT (permissive) · 02c1271594f3092e · report

Tasks

Adversarial Robustness

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Label Smoothing

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