Methods › General › Adversarial Training › AdvProp

AdvProp

6 papers tagged archive 2025-07-28

Introduced by Cihang Xie et al. in Adversarial Examples Improve Image Recognition

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

AdvProp is an adversarial training scheme which treats adversarial examples as additional examples, to prevent overfitting. Key to the method is the usage of a separate auxiliary batch norm for adversarial examples, as they have different underlying distributions to normal examples.

Paper Source

Papers archive 2025-07-28

6 shown of 6, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Improving Model Generalization by On-manifold Adversarial Augmentation in the Frequency Domain
28 Feb 2023 · 0 repositories · arXiv:2302.14302
How explainable are adversarially-robust CNNs?
25 May 2022 · 0 repositories · arXiv:2205.13042
Fast AdvProp
21 Apr 2022 · 1 repository · arXiv:2204.09838

Syntology ran 4 of 4 samples · 0 unverified
Pyramid Adversarial Training Improves ViT Performance
30 Nov 2021 · 1 repository · arXiv:2111.15121
Advanced Graph and Sequence Neural Networks for Molecular Property Prediction and Drug Discovery
2 Dec 2020 · 1 repository · arXiv:2012.01981
Adversarial Examples Improve Image Recognition
21 Nov 2019 · 6 repositories · arXiv:1911.09665

Syntology ran 2 of 2 samples · 0 unverified

Tasks archive 2025-07-28

11 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

Task	Papers
Data Augmentation	3
Domain Generalization	2
Image Classification	2
Adversarial Attack	1
BIG-bench Machine Learning	1
Drug Discovery	1
Feature Importance	1
Molecular Property Prediction	1
Object Detection	1
Property Prediction	1
object-detection	1

Usage over time archive 2025-07-28

Papers per year the archive tags with this method, by the paper's archive date (6 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Adversarial Training

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