AI Paper Atlas

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

Robustness between the worst and average case

Leslie Rice, Anna Bair, Huan Zhang, J Zico Kolter

NeurIPS 2021 · Adversarial robustness and security › Certified and empirical robustness · approach: Provable robust learning guarantees · 0 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
estimate with path sampling + HMC instead of Monte Carlo (q=1000)standard-trained model · MNIST l-inf ballq-norm of loss (higher = better estimate): 0.268 -> 4.456Improved
estimate with path sampling + HMC instead of Monte Carlo (q=1000)standard-trained model · CIFAR-10 l-inf ballq-norm of loss (higher = better estimate): 1.216 -> 4.991Improved
estimate with path sampling + HMC instead of Monte Carlo (q=1000)PGD-50 trained model · MNIST l-inf ballq-norm of loss: 0.081 -> 0.187Improved
estimate with path sampling + HMC instead of Monte Carlo (q=1000)PGD-10 trained model · CIFAR-10 l-inf ballq-norm of loss: 0.761 -> 0.796Improved
estimate with path sampling + HMC instead of Monte Carlo at q=1standard-trained model · MNIST l-inf ballq-norm of loss: 0.043 -> 0.043No clear effect
estimate with path sampling + random-walk Metropolis instead of Monte Carlo (q=1000)standard-trained model · CIFAR-10 spatial transformationsq-norm of loss: 3.865 -> 4.889Improved
estimate with path sampling + random-walk Metropolis instead of Monte Carlo (q=1000)MC q=1 trained model · CIFAR-10 spatial transformationsq-norm of loss: 1.300 -> 1.711Improved
train with path sampling objective (q=1000) instead of Monte Carlo objective (q=1000)MNIST CNN · MNIST l-inf ballPS+HMC q=1000 robust loss (lower is better): 1.268 -> 0.684Improved
train with path sampling objective (q=100) instead of Monte Carlo objective (q=100)MNIST CNN · MNIST l-inf ballPS+HMC q=1000 robust loss (lower is better): 1.261 -> 0.816Improved
train with path sampling objective (q=10) instead of Monte Carlo objective (q=10)MNIST CNN · MNIST l-inf ballPS+HMC q=1000 robust loss (lower is better): 1.336 -> 1.307No clear effect
train with path sampling objective (q=10) instead of Monte Carlo objective (q=10)MNIST CNN · MNIST l-inf balladversarial loss (PGD-100): 3.722 -> 5.012Hurt
train with path sampling objective (q=1000) instead of Monte Carlo objective (q=1000)MNIST CNN · MNIST l-inf balladversarial loss (PGD-100): 3.488 -> 1.646Improved

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