Self-Influence Governs Generalization: A von Mises Expansion Approach

John Sous, Anirvan M. Sengupta

Keywords: Generalization, Influence Functions, Hierarchical Sampling, von Mises Expansion

TL;DR: Generalization is governed by self-influence and same-sample cross-influence arising from hierarchical sampling, while conservative influence estimates signal the onset of memorization.

Abstract: We study generalization through the lens of self-influence: how strongly a training example affects the learned predictor and its associated losses. Using a second-order von Mises expansion of the loss functional, we derive leading-order influence-based estimators for the expected generalization gap. For i.i.d. samples, the expected gap is governed by average self-influence. For hierarchical sampling, additional same-parent cross-influence terms appear, providing a mechanism by which augmentations and correlated views affect generalization. Empirically, the resulting estimators accurately track generalization in well-generalizing regimes and become conservative near memorization, providing a diagnostic signal for the breakdown of the leading-order approximation.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 149