Sign in

Adversarial Robustness Limits via Scaling-Law and Human-Alignment Studies

By Brian Bartoldson and others
This paper revisits the simple, long-studied, yet still unsolved problem of making image classifiers robust to imperceptible perturbations. Taking CIFAR10 as an example, SOTA clean accuracy is about \(100\)%, but SOTA robustness to \(\ell_{\infty}\)-norm bounded perturbations barely exceeds \(70\)%. To understand this gap, we analyze how model size, dataset size,... Show more
July 10, 2024
=
0
Loading PDF…
Loading full text...
Similar articles
Loading recommendations...
=
0
x1
Adversarial Robustness Limits via Scaling-Law and Human-Alignment Studies
Click on play to start listening