A Variational Information Theoretic Approach to Out-of-Distribution Detection

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Mondal, Sudeepta, Jiang, Zhuolin, Sundaramoorthi, Ganesh
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911009045741568
author Mondal, Sudeepta
Jiang, Zhuolin
Sundaramoorthi, Ganesh
author_facet Mondal, Sudeepta
Jiang, Zhuolin
Sundaramoorthi, Ganesh
contents We present a theory for the construction of out-of-distribution (OOD) detection features for neural networks. We introduce random features for OOD through a novel information-theoretic loss functional consisting of two terms, the first based on the KL divergence separates resulting in-distribution (ID) and OOD feature distributions and the second term is the Information Bottleneck, which favors compressed features that retain the OOD information. We formulate a variational procedure to optimize the loss and obtain OOD features. Based on assumptions on OOD distributions, one can recover properties of existing OOD features, i.e., shaping functions. Furthermore, we show that our theory can predict a new shaping function that out-performs existing ones on OOD benchmarks. Our theory provides a general framework for constructing a variety of new features with clear explainability.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14194
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Variational Information Theoretic Approach to Out-of-Distribution Detection
Mondal, Sudeepta
Jiang, Zhuolin
Sundaramoorthi, Ganesh
Machine Learning
We present a theory for the construction of out-of-distribution (OOD) detection features for neural networks. We introduce random features for OOD through a novel information-theoretic loss functional consisting of two terms, the first based on the KL divergence separates resulting in-distribution (ID) and OOD feature distributions and the second term is the Information Bottleneck, which favors compressed features that retain the OOD information. We formulate a variational procedure to optimize the loss and obtain OOD features. Based on assumptions on OOD distributions, one can recover properties of existing OOD features, i.e., shaping functions. Furthermore, we show that our theory can predict a new shaping function that out-performs existing ones on OOD benchmarks. Our theory provides a general framework for constructing a variety of new features with clear explainability.
title A Variational Information Theoretic Approach to Out-of-Distribution Detection
topic Machine Learning
url https://arxiv.org/abs/2506.14194