AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

AdAUC: End-to-end Adversarial AUC Optimization Against Long-tail Problems

Wenzheng Hou, Qianqian Xu, Zhiyong Yang, Shilong Bao, Yuan He, Qingming Huang

ICML 2022 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training refinements · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use AdAUC AT2 rather than the strongest CE-based model for adversarial trainingWideResNet-28 · CIFAR-10-LT and CIFAR-100-LT, PGD-5/10/20, C&W, and AAAUC: CIFAR-10-LT PGD-5: 0.4870 -> 0.5591; CIFAR-100-LT PGD-5: 0.3183 -> 0.4644Improved
Use AdAUC AT2 rather than the strongest CE-based model for adversarial trainingWideResNet-28 · MNIST-LT, PGD-5/10/20, C&W, and AAAUC: PGD-5: 0.8912 -> 0.9436; AA: 0.8818 -> 0.9311Improved
Use AdAUC AT2 rather than the strongest CE-based modelWideResNet-28 · CIFAR-10-LT clean and MNIST-LT cleanAUC: CIFAR-10-LT: 0.6833 -> 0.7528; MNIST-LT: 0.9736 -> 0.9852Improved
Use AdAUC AT2 rather than the strongest CE-based modelWideResNet-28 · CIFAR-100-LT cleanAUC: 0.6382 -> 0.6313Hurt
Add FOSC-based adaptive control of adversarial-example strength (ablation)AdAUC AT1 · CIFAR-10-LT and CIFAR-100-LT, including PGD-20AUC: CIFAR-10-LT PGD-20: 0.4224 -> 0.5283; CIFAR-100-LT PGD-20: 0.2989 -> 0.4065Improved
Add FOSC-based adaptive control of adversarial-example strength (ablation)AdAUC AT1 · MNIST-LTAUC: PGD-10: 0.9395 -> 0.9347; PGD-20: 0.9382 -> 0.9323Mixed
Use AUC optimization instead of CE classification lossnatural training on long-tail datasets · CIFAR-10-LT and MNIST-LT cleanAUC: CIFAR-10-LT: 0.7264 -> 0.7885; MNIST-LT: 0.9736 -> 0.9904Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy