Prototype-based Optimal Transport for Out-of-Distribution Detection

TL;DR

This paper introduces a novel OOD detection method using prototype-based optimal transport to measure distribution discrepancies, enhanced by virtual outliers, resulting in improved accuracy over existing techniques.

Contribution

The paper proposes a new OOD detection approach that combines optimal transport with virtual outliers to better distinguish ID and OOD data, especially near the ID boundary.

Findings

Outperforms state-of-the-art OOD detection methods

Effectively identifies OOD inputs close to ID data

Enhances detection accuracy through virtual outlier generation

Abstract

Detecting Out-of-Distribution (OOD) inputs is crucial for improving the reliability of deep neural networks in the real-world deployment. In this paper, inspired by the inherent distribution shift between ID and OOD data, we propose a novel method that leverages optimal transport to measure the distribution discrepancy between test inputs and ID prototypes. The resulting transport costs are used to quantify the individual contribution of each test input to the overall discrepancy, serving as a desirable measure for OOD detection. To address the issue that solely relying on the transport costs to ID prototypes is inadequate for identifying OOD inputs closer to ID data, we generate virtual outliers to approximate the OOD region via linear extrapolation. By combining the transport costs to ID prototypes with the costs to virtual outliers, the detection of OOD data near ID data is emphasized, thereby enhancing the distinction between ID and OOD inputs. Experiments demonstrate the superiority of our method over state-of-the-art methods.