Saved in:
Bibliographic Details
Main Authors: Mei, Yihan, Wang, Xinyu, Zhang, Dell, Wang, Xiaoling
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.04759
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929340400271360
author Mei, Yihan
Wang, Xinyu
Zhang, Dell
Wang, Xiaoling
author_facet Mei, Yihan
Wang, Xinyu
Zhang, Dell
Wang, Xiaoling
contents In today's interconnected world, achieving reliable out-of-distribution (OOD) detection poses a significant challenge for machine learning models. While numerous studies have introduced improved approaches for multi-class OOD detection tasks, the investigation into multi-label OOD detection tasks has been notably limited. We introduce Spectral Normalized Joint Energy (SNoJoE), a method that consolidates label-specific information across multiple labels through the theoretically justified concept of an energy-based function. Throughout the training process, we employ spectral normalization to manage the model's feature space, thereby enhancing model efficacy and generalization, in addition to bolstering robustness. Our findings indicate that the application of spectral normalization to joint energy scores notably amplifies the model's capability for OOD detection. We perform OOD detection experiments utilizing PASCAL-VOC as the in-distribution dataset and ImageNet-22K or Texture as the out-of-distribution datasets. Our experimental results reveal that, in comparison to prior top performances, SNoJoE achieves 11% and 54% relative reductions in FPR95 on the respective OOD datasets, thereby defining the new state of the art in this field of study.
format Preprint
id arxiv_https___arxiv_org_abs_2405_04759
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-Label Out-of-Distribution Detection with Spectral Normalized Joint Energy
Mei, Yihan
Wang, Xinyu
Zhang, Dell
Wang, Xiaoling
Computer Vision and Pattern Recognition
Machine Learning
In today's interconnected world, achieving reliable out-of-distribution (OOD) detection poses a significant challenge for machine learning models. While numerous studies have introduced improved approaches for multi-class OOD detection tasks, the investigation into multi-label OOD detection tasks has been notably limited. We introduce Spectral Normalized Joint Energy (SNoJoE), a method that consolidates label-specific information across multiple labels through the theoretically justified concept of an energy-based function. Throughout the training process, we employ spectral normalization to manage the model's feature space, thereby enhancing model efficacy and generalization, in addition to bolstering robustness. Our findings indicate that the application of spectral normalization to joint energy scores notably amplifies the model's capability for OOD detection. We perform OOD detection experiments utilizing PASCAL-VOC as the in-distribution dataset and ImageNet-22K or Texture as the out-of-distribution datasets. Our experimental results reveal that, in comparison to prior top performances, SNoJoE achieves 11% and 54% relative reductions in FPR95 on the respective OOD datasets, thereby defining the new state of the art in this field of study.
title Multi-Label Out-of-Distribution Detection with Spectral Normalized Joint Energy
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2405.04759