On a wide selection of challenging vision, language, and biology OOD benchmarks, it is shown that RMD meaningfully improves upon MD performance (by up to 15% AUROC on genomics OOD).
Mahalanobis distance (MD) is a simple and popular post-processing method for detecting out-of-distribution (OOD) inputs in neural networks. We analyze its failure modes for near-OOD detection and propose a simple fix called relative Mahalanobis distance (RMD) which improves performance and is more robust to hyperparameter choice. On a wide selection of challenging vision, language, and biology OOD benchmarks (CIFAR-100 vs CIFAR-10, CLINC OOD intent detection, Genomics OOD), we show that RMD meaningfully improves upon MD performance (by up to 15% AUROC on genomics OOD).
Stanislav Fort
3 papers
J. Liu
2 papers
Shreyas Padhy
3 papers