MLITLGOct 30, 2024

Generalization Bounds via Conditional $f$-Information

arXiv:2410.22887v14 citationsh-index: 8NIPS
Originality Incremental advance
AI Analysis

This work addresses the fundamental problem of understanding generalization in machine learning, though it appears incremental as it builds on existing information-theoretic frameworks.

The authors tackled the problem of deriving generalization bounds in machine learning by introducing novel information-theoretic bounds using conditional f-information, applicable to both bounded and unbounded loss functions. Their approach recovers previous results and empirically demonstrates improvements over existing bounds.

In this work, we introduce novel information-theoretic generalization bounds using the conditional $f$-information framework, an extension of the traditional conditional mutual information (MI) framework. We provide a generic approach to derive generalization bounds via $f$-information in the supersample setting, applicable to both bounded and unbounded loss functions. Unlike previous MI-based bounds, our proof strategy does not rely on upper bounding the cumulant-generating function (CGF) in the variational formula of MI. Instead, we set the CGF or its upper bound to zero by carefully selecting the measurable function invoked in the variational formula. Although some of our techniques are partially inspired by recent advances in the coin-betting framework (e.g., Jang et al. (2023)), our results are independent of any previous findings from regret guarantees of online gambling algorithms. Additionally, our newly derived MI-based bound recovers many previous results and improves our understanding of their potential limitations. Finally, we empirically compare various $f$-information measures for generalization, demonstrating the improvement of our new bounds over the previous bounds.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes