Comparison theorems on large-margin learning
Machine Learning
2019-08-14 v1 Machine Learning
Abstract
This paper studies binary classification problem associated with a family of loss functions called large-margin unified machines (LUM), which offers a natural bridge between distribution-based likelihood approaches and margin-based approaches. It also can overcome the so-called data piling issue of support vector machine in the high-dimension and low-sample size setting. In this paper we establish some new comparison theorems for all LUM loss functions which play a key role in the further error analysis of large-margin learning algorithms.
Cite
@article{arxiv.1908.04470,
title = {Comparison theorems on large-margin learning},
author = {Jun Fan and Dao-Hong Xiang},
journal= {arXiv preprint arXiv:1908.04470},
year = {2019}
}