Tight Sample Complexity of Large-Margin Learning
Machine Learning
2015-03-17 v2 Probability
Statistics Theory
Machine Learning
Statistics Theory
Abstract
We obtain a tight distribution-specific characterization of the sample complexity of large-margin classification with L_2 regularization: We introduce the \gamma-adapted-dimension, which is a simple function of the spectrum of a distribution's covariance matrix, and show distribution-specific upper and lower bounds on the sample complexity, both governed by the \gamma-adapted-dimension of the source distribution. We conclude that this new quantity tightly characterizes the true sample complexity of large-margin classification. The bounds hold for a rich family of sub-Gaussian distributions.
Keywords
Cite
@article{arxiv.1011.5053,
title = {Tight Sample Complexity of Large-Margin Learning},
author = {Sivan Sabato and Nathan Srebro and Naftali Tishby},
journal= {arXiv preprint arXiv:1011.5053},
year = {2015}
}
Comments
Appearing in Neural Information Processing Systems (NIPS) 2010; This is the full version, including appendix with proofs; Also with some corrections