Algorithmic stability and hypothesis complexity
Machine Learning
2017-08-04 v2 Machine Learning
Abstract
We introduce a notion of algorithmic stability of learning algorithms---that we term \emph{argument stability}---that captures stability of the hypothesis output by the learning algorithm in the normed space of functions from which hypotheses are selected. The main result of the paper bounds the generalization error of any learning algorithm in terms of its argument stability. The bounds are based on martingale inequalities in the Banach space to which the hypotheses belong. We apply the general bounds to bound the performance of some learning algorithms based on empirical risk minimization and stochastic gradient descent.
Cite
@article{arxiv.1702.08712,
title = {Algorithmic stability and hypothesis complexity},
author = {Tongliang Liu and Gábor Lugosi and Gergely Neu and Dacheng Tao},
journal= {arXiv preprint arXiv:1702.08712},
year = {2017}
}