Approximate Description Length, Covering Numbers, and VC Dimension
Machine Learning
2022-09-27 v1 Data Structures and Algorithms
Machine Learning
Abstract
Recently, Daniely and Granot [arXiv:1910.05697] introduced a new notion of complexity called Approximate Description Length (ADL). They used it to derive novel generalization bounds for neural networks, that despite substantial work, were out of reach for more classical techniques such as discretization, Covering Numbers and Rademacher Complexity. In this paper we explore how ADL relates to classical notions of function complexity such as Covering Numbers and VC Dimension. We find that for functions whose range is the reals, ADL is essentially equivalent to these classical complexity measures. However, this equivalence breaks for functions with high dimensional range.
Cite
@article{arxiv.2209.12882,
title = {Approximate Description Length, Covering Numbers, and VC Dimension},
author = {Amit Daniely and Gal Katzhendler},
journal= {arXiv preprint arXiv:2209.12882},
year = {2022}
}