AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

Disrupting Deep Uncertainty Estimation Without Harming Accuracy

Ido Galil, Ran El-Yaniv

NeurIPS 2021 · Adversarial robustness and security › Black-box and transfer attacks · approach: Robustness benchmarks and defenses · 4 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
Add ACE black-box perturbations with ε=0.05 (ablation)ResNet50 softmax confidence · ImageNet validationAURC: 69.9 -> 456.6Hurt
Add ACE black-box perturbations with ε=0.05 (ablation)ResNet50 softmax classifier · ImageNet validationAccuracy: 76.01 -> 76.01No clear effect
Use white-box access rather than a black-box proxy (ablation)EfficientNetB0 softmax uncertainty under ACE · ImageNet validationAURC: 69.1 -> 553.3Hurt
Use a matching ResNet50 proxy rather than a foreign-architecture proxy (ablation)ACE attacking a ResNet50 ensemble · ImageNet validationAURC: 290.8 -> 381Hurt
Increase the ensemble size from 3 to 10 (ablation)ResNet50 ensemble under white-box ACE · ImageNet validationAURC: 533.4 -> 510.5No clear effect
Attack MC-Dropout using ACE with ε=0.05 (ablation)MobileNetV2 MC-Dropout, 30 passes · ImageNet validationAURC: 94.5 -> 275.2Hurt
Increase MC-Dropout passes from 10 to 30 under black-box ACE (ablation)MobileNetV2 MC-Dropout · ImageNet validationAURC: 275.3 -> 275.2No clear effect
Target softmax indirectly instead of targeting predictive entropy directly (ablation)White-box ACE on MC-Dropout · ImageNet validationAURC: 587.6 -> 622.9; 433.9 -> 502.9Mixed
Increase MC-Dropout passes from 10 to 30 for direct attacks (ablation)White-box ACE targeting predictive entropy · ImageNet validationAURC: 587.9 -> 587.6; 521.2 -> 433.9No clear effect
Attack softmax rather than variance directly (ablation)MC-Dropout epistemic uncertainty · ImageNet validationAURC: 624.3 vs 453.2; 565.9 vs 414.5Mixed
Add ACE directly to the SelectiveNet selector head at ε=0.005 (ablation)SelectiveNet with ResNet18 backbone · CIFAR-10 testSelective risk: 0.0012 -> 0.0747Hurt
Attack softmax indirectly rather than the selector directly (ablation)SelectiveNet with VGG16 backbone, ε=0.005 · CIFAR-10 testSelective risk: 0.0772 -> 0.0829Hurt

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