Multi-category classifiers and sample width

Anthony, Martin and Ratsaby, Joel (2016) Multi-category classifiers and sample width. Journal of Computer and System Sciences, 82 (8). pp. 1223-1231. ISSN 0022-0000

PDF - Accepted Version Restricted to Repository staff only until 23 June 2017. Download (390Kb)

• LSE expert's profile

• Published item via DOI

Abstract

In a recent paper, the authors introduced the notion of sample width for binary classifier defined on the set of real numbers. It was shown that the performance of such classifier could be quantified in terms of this sample width. This paper considers how to adapt the idea of sample width so that it can be applied in cases where the classifier are multi-category and are defined on some arbitrary metric space.

Item Type:	Article
Official URL:	http://www.journals.elsevier.com/journal-of-comput...
Additional Information:	© 2016 Published by Elsevier Inc.
Library of Congress subject classification:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Sets:	Departments > Mathematics
Project and Funder Information:	Project ID Funder Name Funder ID ST-2007-216886 ST Programme of the European Community UNSPECIFIED UNSPECIFIED Suntory and Toyota International Centres for Economics and Related Disciplines UNSPECIFIED
Date Deposited:	29 Apr 2016 11:20
URL:	http://eprints.lse.ac.uk/66277/

Actions (login required)

Record administration - authorised staff only