OpenOOD: Benchmarking Generalized Out-of-Distribution Detection

TL;DR

OpenOOD provides a unified benchmark and codebase for generalized out-of-distribution detection, enabling fair comparisons and advancing the field by evaluating over 30 methods across related domains.

Contribution

It introduces a comprehensive, unified benchmark and codebase for OOD detection, integrating methods from related fields under a generalized framework.

Findings

Significant progress in OOD detection over recent years.

Preprocessing and post-hoc methods show strong potential.

The benchmark facilitates fairer comparisons across methods.

Abstract

Out-of-distribution (OOD) detection is vital to safety-critical machine learning applications and has thus been extensively studied, with a plethora of methods developed in the literature. However, the field currently lacks a unified, strictly formulated, and comprehensive benchmark, which often results in unfair comparisons and inconclusive results. From the problem setting perspective, OOD detection is closely related to neighboring fields including anomaly detection (AD), open set recognition (OSR), and model uncertainty, since methods developed for one domain are often applicable to each other. To help the community to improve the evaluation and advance, we build a unified, well-structured codebase called OpenOOD, which implements over 30 methods developed in relevant fields and provides a comprehensive benchmark under the recently proposed generalized OOD detection framework. With a comprehensive comparison of these methods, we are gratified that the field has progressed significantly over the past few years, where both preprocessing methods and the orthogonal post-hoc methods show strong potential.