Comprehensive OOD Detection Improvements

Files

2401.10176.pdf (146.59 KB)

Links to Files

https://arxiv.org/abs/2401.10176

Permanent Link

https://doi.org/10.48550/arXiv.2401.10176
 http://hdl.handle.net/11603/31568

Collections

UMBC Data Science
UMBC Faculty Collection

Author/Creator

Author/Creator ORCID

https://orcid.org/0000-0002-9900-1972

Date

2024-01-18

Type of Work

journal articles
preprints
Text

Department

Program

Citation of Original Publication

Rights

This work was written as part of one of the author's official duties as an Employee of the United States Government and is therefore a work of the United States Government. In accordance with 17 U.S.C. 105, no copyright protection is available for such works under U.S. Law.
Public Domain Mark 1.0 Universal

Subjects

Abstract

As machine learning becomes increasingly prevalent in impactful decisions, recognizing when inference data is outside the model's expected input distribution is paramount for giving context to predictions. Out-of-distribution (OOD) detection methods have been created for this task. Such methods can be split into representation-based or logit-based methods from whether they respectively utilize the model's embeddings or predictions for OOD detection. In contrast to most papers which solely focus on one such group, we address both. We employ dimensionality reduction on feature embeddings in representation-based methods for both time speedups and improved performance. Additionally, we propose DICE-COL, a modification of the popular logit-based method Directed Sparsification (DICE) that resolves an unnoticed flaw. We demonstrate the effectiveness of our methods on the OpenOODv1.5 benchmark framework, where they significantly improve performance and set state-of-the-art results.