A Combinatorial Characterization of Supervised Online Learnability
Machine Learning
2024-02-12 v2
Abstract
We study the online learnability of hypothesis classes with respect to arbitrary, but bounded loss functions. No characterization of online learnability is known at this level of generality. We give a new scale-sensitive combinatorial dimension, named the sequential minimax dimension, and show that it gives a tight quantitative characterization of online learnability. In addition, we show that the sequential minimax dimension subsumes most existing combinatorial dimensions in online learning theory.
Cite
@article{arxiv.2307.03816,
title = {A Combinatorial Characterization of Supervised Online Learnability},
author = {Vinod Raman and Unique Subedi and Ambuj Tewari},
journal= {arXiv preprint arXiv:2307.03816},
year = {2024}
}
Comments
20 pages. arXiv admin note: text overlap with arXiv:2306.06247