AI Paper Atlas

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

The Adversarial Attack and Detection under the Fisher Information Metric

Chenxiao Zhao, P. Thomas Fletcher, Mixue Yu, Yaxin Peng, Guixu Zhang, Chaomin Shen

AAAI 2019 · Adversarial robustness and security › Certified and empirical robustness · approach: Attack evaluation and defense critique · 42 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
Replace the input gradient direction with the top Fisher information eigenvectorOne-step attack on LeNet-5 · MNISTmean least ℓ2 perturbation norm: 2.11 -> 1.80Improved
Replace the input gradient direction with the top Fisher information eigenvectorOne-step attack on VGG · CIFAR-10fooling rate (%): 95.21 -> 97.85Improved
Use Fisher information eigenvalues as detector features instead of combined kernel density and Bayesian uncertainty featuresRandom-forest adversarial detector · CIFAR-10, Opt attacksAUC (%): 93.77 -> 99.45Improved
Use Fisher information eigenvalues as detector features instead of combined kernel density and Bayesian uncertainty featuresRandom-forest adversarial detector · CIFAR-10, OTCM attacksAUC (%): 94.45 -> 93.68Hurt
Use Fisher information eigenvalues as detector features instead of combined kernel density and Bayesian uncertainty featuresRandom-forest adversarial detector · CIFAR-10, OSSA attacksAUC (%): 93.54 -> 98.01Improved
Train the detector on OSSA rather than other attack typesRandom-forest detector · MNIST cross-attack generalizationAUC (%): 89.47Hurt
Use alias sampling with power iteration instead of exact eigen-decompositionOSSA computation · CIFAR-10time (seconds): 1.49±{0.12} -> 0.25±{0.01}Improved

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