Nearest Neighbor Guidance for Out-of-Distribution Detection

TL;DR

This paper introduces NNGuide, a method that improves out-of-distribution detection by guiding classifier scores to better respect data boundary geometry, reducing overconfidence in OOD samples and achieving state-of-the-art results.

Contribution

The paper proposes NNGuide, a novel approach that enhances classifier-based OOD detection by incorporating nearest neighbor guidance to respect data manifold boundaries.

Findings

NNGuide significantly improves OOD detection performance.

Achieves state-of-the-art AUROC, FPR95, and AUPR metrics.

Effective under diverse settings, including distribution shifts.

Abstract

Detecting out-of-distribution (OOD) samples are crucial for machine learning models deployed in open-world environments. Classifier-based scores are a standard approach for OOD detection due to their fine-grained detection capability. However, these scores often suffer from overconfidence issues, misclassifying OOD samples distant from the in-distribution region. To address this challenge, we propose a method called Nearest Neighbor Guidance (NNGuide) that guides the classifier-based score to respect the boundary geometry of the data manifold. NNGuide reduces the overconfidence of OOD samples while preserving the fine-grained capability of the classifier-based score. We conduct extensive experiments on ImageNet OOD detection benchmarks under diverse settings, including a scenario where the ID data undergoes natural distribution shift. Our results demonstrate that NNGuide provides a significant performance improvement on the base detection scores, achieving state-of-the-art results on both AUROC, FPR95, and AUPR metrics. The code is given at \url{https://github.com/roomo7time/nnguide}.