Out-Of-Distribution Detection with Diversification (Provably)

TL;DR

This paper introduces diverseMix, a method that enhances out-of-distribution detection by increasing auxiliary outlier diversity, backed by theoretical guarantees and extensive experimental validation on large-scale benchmarks.

Contribution

The paper proposes diverseMix, a practical approach with theoretical guarantees, to improve OOD detection by diversifying auxiliary outliers efficiently.

Findings

diverseMix outperforms existing methods on large-scale benchmarks

Increasing auxiliary outlier diversity improves OOD detection accuracy

Theoretical analysis confirms the effectiveness of diversity in outlier sets

Abstract

Out-of-distribution (OOD) detection is crucial for ensuring reliable deployment of machine learning models. Recent advancements focus on utilizing easily accessible auxiliary outliers (e.g., data from the web or other datasets) in training. However, we experimentally reveal that these methods still struggle to generalize their detection capabilities to unknown OOD data, due to the limited diversity of the auxiliary outliers collected. Therefore, we thoroughly examine this problem from the generalization perspective and demonstrate that a more diverse set of auxiliary outliers is essential for enhancing the detection capabilities. However, in practice, it is difficult and costly to collect sufficiently diverse auxiliary outlier data. Therefore, we propose a simple yet practical approach with a theoretical guarantee, termed Diversity-induced Mixup for OOD detection (diverseMix), which enhances the diversity of auxiliary outlier set for training in an efficient way. Extensive experiments show that diverseMix achieves superior performance on commonly used and recent challenging large-scale benchmarks, which further confirm the importance of the diversity of auxiliary outliers.