Class-Size Independent Generalization Analsysis of Some Discriminative Multi-Category Classification

Abstract

We consider the problem of deriving class-size independent generaliza- tion bounds for some regularized discriminative multi-category classi- fication methods. In particular, we obtain an expected generalization bound for a standard formulation of multi-category support vector ma- chines. Based on the theoretical result, we argue that the formula- tion over-penalizes misclassification error, which in theory may lead to poor generalization performance. A remedy, based on a generalization of multi-category logistic regression (conditional maximum entropy), is then proposed, and its theoretical properties are examined.

Cite

Text

Zhang. "Class-Size Independent Generalization Analsysis of Some Discriminative Multi-Category Classification." Neural Information Processing Systems, 2004.

Markdown

[Zhang. "Class-Size Independent Generalization Analsysis of Some Discriminative Multi-Category Classification." Neural Information Processing Systems, 2004.](https://mlanthology.org/neurips/2004/zhang2004neurips-classsize/)

BibTeX

@inproceedings{zhang2004neurips-classsize,
  title     = {{Class-Size Independent Generalization Analsysis of Some Discriminative Multi-Category Classification}},
  author    = {Zhang, Tong},
  booktitle = {Neural Information Processing Systems},
  year      = {2004},
  pages     = {1625-1632},
  url       = {https://mlanthology.org/neurips/2004/zhang2004neurips-classsize/}
}