Email this article Tight risk bounds for multi-class margin classifiers
- Authors: Maximov Y.1, Reshetova D.2,3
-
Affiliations:
- Skolkovo Institute of Science and Technology Skolkovo Innovation Center
- Predictive Modeling and Optimization Department Institute of Information Transmission Problems
- Predictive Modeling and Optimization Laboratory Moscow Institute of Physics and Technology
- Issue: Vol 26, No 4 (2016)
- Pages: 673-680
- Section: Mathematical Method in Pattern Recognition
- URL: https://journals.rcsi.science/1054-6618/article/view/194905
- DOI: https://doi.org/10.1134/S105466181604009X
- ID: 194905
Cite item
Open Access
Access granted
Subscription Access
Abstract
We consider a problem of risk estimation for large-margin multi-class classifiers. We propose a novel risk bound for the multi-class classification problem. The bound involves the marginal distribution of the classifier and the Rademacher complexity of the hypothesis class. We prove that our bound is tight in the number of classes. Finally, we compare our bound with the related ones and provide a simplified version of the bound for the multi-class classification with kernel based hypotheses.
About the authors
Yu. Maximov
Skolkovo Institute of Science and Technology Skolkovo Innovation Center
Author for correspondence.
Email: yurymaximov@iitp.ru
Russian Federation, Building 3, Moscow, 143026
D. Reshetova
Predictive Modeling and Optimization Department Institute of Information Transmission Problems; Predictive Modeling and Optimization Laboratory Moscow Institute of Physics and Technology
Email: yurymaximov@iitp.ru
Russian Federation, Bol’shoi Karetnyi 19/1, Moscow, 127051; Kerchenskaya ul. 1a/1, Moscow, 117303
Supplementary files
