A Margin-based Multiclass Generalization Bound via Geometric Complexity
Abstract
There has been considerable effort to better understand the generalization capabilities of deep neural networks both as a means to unlock a theoretical understanding of their success as well as providing directions for further improvements. In this paper, we investigate margin-based multiclass generalization bounds for neural networks which rely on a recent complexity measure, the geometric complexity, developed for neural networks. We derive a new upper bound on the generalization error which scales with the margin-normalized geometric complexity of the network and which holds for a broad family of data distributions and model classes. Our generalization bound is empirically investigated for a ResNet-18 model trained with SGD on the CIFAR-10 and CIFAR-100 datasets with both original and random labels.
Cite
@article{arxiv.2405.18590,
title = {A Margin-based Multiclass Generalization Bound via Geometric Complexity},
author = {Michael Munn and Benoit Dherin and Javier Gonzalvo},
journal= {arXiv preprint arXiv:2405.18590},
year = {2024}
}
Comments
Accepted as an ICML 2023 workshop paper (Topology, Algebra and Geometry in Machine Learning)