RankFeat: Rank-1 Feature Removal for Out-of-distribution Detection

Yue Song, Nicu Sebe, Wei Wang

Keywords: out-of-distribution detection, distribution shifts

TL;DR: We propose a simple yet effective post hoc OOD detection method by removing the rank-1 matrix from the high-level feature.

Abstract: The task of out-of-distribution (OOD) detection is crucial for deploying machine learning models in real-world settings. In this paper, we observe that the singular value distributions of the in-distribution (ID) and OOD features are quite different: the OOD feature matrix tends to have a larger dominant singular value than the ID feature, and the class predictions of OOD samples are largely determined by it. This observation motivates us to propose RankFeat, a simple yet effective post hoc approach for OOD detection by removing the rank-1 matrix composed of the largest singular value and the associated singular vectors from the high-level feature. RankFeat achieves state-of-the-art performance and reduces the average false positive rate (FPR95) by 17.90% compared with the previous best method. Extensive ablation studies and comprehensive theoretical analyses are presented to support the empirical results.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2209.08590/code)