Saved in:
Bibliographic Details
Main Authors: Liu, Wenliang, Yu, Guanding, Wang, Lele, Liao, Renjie
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.19895
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915062333046784
author Liu, Wenliang
Yu, Guanding
Wang, Lele
Liao, Renjie
author_facet Liu, Wenliang
Yu, Guanding
Wang, Lele
Liao, Renjie
contents We study the Out-of-Distribution (OOD) generalization in machine learning and propose a general framework that establishes information-theoretic generalization bounds. Our framework interpolates freely between Integral Probability Metric (IPM) and $f$-divergence, which naturally recovers some known results (including Wasserstein- and KL-bounds), as well as yields new generalization bounds. Additionally, we show that our framework admits an optimal transport interpretation. When evaluated in two concrete examples, the proposed bounds either strictly improve upon existing bounds in some cases or match the best existing OOD generalization bounds. Moreover, by focusing on $f$-divergence and combining it with the Conditional Mutual Information (CMI) methods, we derive a family of CMI-based generalization bounds, which include the state-of-the-art ICIMI bound as a special instance. Finally, leveraging these findings, we analyze the generalization of the Stochastic Gradient Langevin Dynamics (SGLD) algorithm, showing that our derived generalization bounds outperform existing information-theoretic generalization bounds in certain scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2403_19895
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle An Information-Theoretic Framework for Out-of-Distribution Generalization with Applications to Stochastic Gradient Langevin Dynamics
Liu, Wenliang
Yu, Guanding
Wang, Lele
Liao, Renjie
Information Theory
Machine Learning
We study the Out-of-Distribution (OOD) generalization in machine learning and propose a general framework that establishes information-theoretic generalization bounds. Our framework interpolates freely between Integral Probability Metric (IPM) and $f$-divergence, which naturally recovers some known results (including Wasserstein- and KL-bounds), as well as yields new generalization bounds. Additionally, we show that our framework admits an optimal transport interpretation. When evaluated in two concrete examples, the proposed bounds either strictly improve upon existing bounds in some cases or match the best existing OOD generalization bounds. Moreover, by focusing on $f$-divergence and combining it with the Conditional Mutual Information (CMI) methods, we derive a family of CMI-based generalization bounds, which include the state-of-the-art ICIMI bound as a special instance. Finally, leveraging these findings, we analyze the generalization of the Stochastic Gradient Langevin Dynamics (SGLD) algorithm, showing that our derived generalization bounds outperform existing information-theoretic generalization bounds in certain scenarios.
title An Information-Theoretic Framework for Out-of-Distribution Generalization with Applications to Stochastic Gradient Langevin Dynamics
topic Information Theory
Machine Learning
url https://arxiv.org/abs/2403.19895