← Search

Michael W. Spratling

2 accepted papers

2024

OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift

ICML 2024poster

Existing works have made great progress in improving adversarial robustness, but typically test their method only on data from the same distribution as the training data, i.e. in-distribution (ID) testing. As a result, it is unclear how such robustness generalizes under input distribution shifts, i.…