Balanced Energy Regularization Loss for Out-of-distribution Detection

TL;DR

This paper introduces a balanced energy regularization loss that accounts for class imbalance in auxiliary out-of-distribution data, improving OOD detection across multiple tasks and achieving state-of-the-art results.

Contribution

It proposes a simple, effective method that uses class-wise prior probabilities to better handle class imbalance in OOD detection tasks.

Findings

Outperforms previous energy regularization loss in various tasks.

Achieves state-of-the-art in semantic segmentation and long-tailed classification.

Improves OOD detection accuracy by addressing class imbalance.

Abstract

In the field of out-of-distribution (OOD) detection, a previous method that use auxiliary data as OOD data has shown promising performance. However, the method provides an equal loss to all auxiliary data to differentiate them from inliers. However, based on our observation, in various tasks, there is a general imbalance in the distribution of the auxiliary OOD data across classes. We propose a balanced energy regularization loss that is simple but generally effective for a variety of tasks. Our balanced energy regularization loss utilizes class-wise different prior probabilities for auxiliary data to address the class imbalance in OOD data. The main concept is to regularize auxiliary samples from majority classes, more heavily than those from minority classes. Our approach performs better for OOD detection in semantic segmentation, long-tailed image classification, and image classification than the prior energy regularization loss. Furthermore, our approach achieves state-of-the-art performance in two tasks: OOD detection in semantic segmentation and long-tailed image classification. Code is available at https://github.com/hyunjunChhoi/Balanced_Energy.