AI Paper Atlas

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

LOT: Layer-wise Orthogonal Training on Improving l2 Certified Robustness

Xiaojun Xu, Linyi Li, Bo Li

NeurIPS 2022 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training recipes · 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
replace SOC with LOT layerLipConvnet-20 · CIFAR-10 certified acc rho=108/255certified accuracy: 36.04% -> 36.34%Improved
replace SOC with LOT layer at depth 40LipConvnet-40 · CIFAR-10 vanilla accuracyaccuracy: 72.59% -> 76.75%Improved
replace SOC with LOT at depth 40, certified rho=36/255LipConvnet-40 · CIFAR-10certified accuracy: 59.04% -> 63.50%Improved
replace SOC with LOTLipConvnet-20 · CIFAR-100 vanilla accuracyaccuracy: 47.84% -> 48.81%Improved
LOT pre-computes Newton iteration, replacing SOC exponentialLipConvnet-40 · evaluation timeseconds: 8.668 -> 3.595Improved
add residual connection to LOT (ablation)LOT LipConvnet-20 · CIFAR-10 certified acc rho=108/255certified accuracy: 30.95% -> 32.12%Improved
add residual connection to SOC (ablation)SOC LipConvnet-20 · CIFAR-10 certified acc rho=108/255certified accuracy: 31.08% -> 30.29%No clear effect
zero padding via pre-processing instead of circular padding (ablation)LOT LipConvnet-20 · CIFAR-10 certified acc rho=108/255certified accuracy: 28.78% -> 32.12%Improved
LOT vs SOC under l_inf PGD, depth 40LipConvnet-40 · CIFAR-10 robust accuracyrobust accuracy: 11.68% -> 28.67%Improved
LOT vs SOC under l_inf PGD, depth 5LipConvnet-5 · CIFAR-10 robust accuracyrobust accuracy: 27.18% -> 27.19%No clear effect
add 500K unlabelled semi-supervised data, large radiusLOT LipConvnet-20 · CIFAR-10 certified acc rho=108/255certified accuracy: 36.34% -> 42.39%Improved
add semi-supervised data, vanilla accuracyLOT LipConvnet-20 · CIFAR-10vanilla accuracy: 77.12% -> 71.86%Hurt

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