English

Multi-class SVMs: From Tighter Data-Dependent Generalization Bounds to Novel Algorithms

Machine Learning 2015-06-16 v1

Abstract

This paper studies the generalization performance of multi-class classification algorithms, for which we obtain, for the first time, a data-dependent generalization error bound with a logarithmic dependence on the class size, substantially improving the state-of-the-art linear dependence in the existing data-dependent generalization analysis. The theoretical analysis motivates us to introduce a new multi-class classification machine based on p\ell_p-norm regularization, where the parameter pp controls the complexity of the corresponding bounds. We derive an efficient optimization algorithm based on Fenchel duality theory. Benchmarks on several real-world datasets show that the proposed algorithm can achieve significant accuracy gains over the state of the art.

Keywords

Cite

@article{arxiv.1506.04359,
  title  = {Multi-class SVMs: From Tighter Data-Dependent Generalization Bounds to Novel Algorithms},
  author = {Yunwen Lei and Ürün Dogan and Alexander Binder and Marius Kloft},
  journal= {arXiv preprint arXiv:1506.04359},
  year   = {2015}
}
R2 v1 2026-06-22T09:53:17.209Z