AI Paper Atlas

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

On the Tightness of Semidefinite Relaxations for Certifying Robustness to Adversarial Examples

Richard Zhang

NeurIPS 2020 · Adversarial robustness and security › Certified and empirical robustness · approach: Neural network robustness verification · 3 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
Increase network depth from one layer to two or more layersCNN with the SDP relaxation solved by BM2 · CNN verification instancesnumerical rank-1 success rate: 100% -> 0%Hurt
Use an intermediate radius ρ in the stated loose intervalSDP relaxation for the original adversarial attack problem on dense-1 · dense-1SDP tightness: 0.4≤ρ≤60Hurt
Use a radius in either reported tight intervalSDP relaxation for the original adversarial attack problem on dense-1 · dense-1SDP tightness: 0.07≤ρ≤0.4 and ρ≥60Improved
Use a radius below the reported feasibility thresholdSDP relaxation for the original adversarial attack problem on dense-1 · dense-1SDP feasibility: ρ<0.07Hurt

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