Out of Distribution Detection on ImageNet-O

TL;DR

This paper benchmarks various out-of-distribution detection methods on the new ImageNet-O dataset, highlighting their effectiveness across different models and settings to improve robustness of ImageNet-trained neural networks.

Contribution

It provides the first comprehensive comparison of OOD detection techniques on the novel ImageNet-O dataset, covering multiple models and approaches.

Findings

Deep generative methods show promise in OOD detection.

Performance varies significantly across model architectures.

Access to OOD data influences detection effectiveness.

Abstract

Out of distribution (OOD) detection is a crucial part of making machine learning systems robust. The ImageNet-O dataset is an important tool in testing the robustness of ImageNet trained deep neural networks that are widely used across a variety of systems and applications. We aim to perform a comparative analysis of OOD detection methods on ImageNet-O, a first of its kind dataset with a label distribution different than that of ImageNet, that has been created to aid research in OOD detection for ImageNet models. As this dataset is fairly new, we aim to provide a comprehensive benchmarking of some of the current state of the art OOD detection methods on this novel dataset. This benchmarking covers a variety of model architectures, settings where we haves prior access to the OOD data versus when we don't, predictive score based approaches, deep generative approaches to OOD detection, and more.