Multiclass Learning at One-class Complexity
Multiclass Learning at One-class Complexity
We show in this paper the multiclass classification problem can be implemented in the maximum margin framework with the complexity of one binary Support Vector Machine. We show reducing the complexity does not involve diminishing performance but in some cases this approach can improve the classification accuracy. The multiclass classification is realized in the framework where the output labels are vector valued.
Maximum margin learning, Hilbertian vector label, multiclass learning, Support vector machine
Szedmak, Sandor
c6a84aa3-2956-4acf-8293-a1b676f6d7d8
Shawe-Taylor, John
b1931d97-fdd0-4bc1-89bc-ec01648e928b
2005
Szedmak, Sandor
c6a84aa3-2956-4acf-8293-a1b676f6d7d8
Shawe-Taylor, John
b1931d97-fdd0-4bc1-89bc-ec01648e928b
Szedmak, Sandor and Shawe-Taylor, John
(2005)
Multiclass Learning at One-class Complexity
Record type:
Monograph
(Project Report)
Abstract
We show in this paper the multiclass classification problem can be implemented in the maximum margin framework with the complexity of one binary Support Vector Machine. We show reducing the complexity does not involve diminishing performance but in some cases this approach can improve the classification accuracy. The multiclass classification is realized in the framework where the output labels are vector valued.
More information
Published date: 2005
Keywords:
Maximum margin learning, Hilbertian vector label, multiclass learning, Support vector machine
Organisations:
Electronics & Computer Science
Identifiers
Local EPrints ID: 261157
URI: http://eprints.soton.ac.uk/id/eprint/261157
PURE UUID: dea3b36d-14b6-4627-b08c-be4064f0cfa3
Catalogue record
Date deposited: 19 Aug 2005
Last modified: 14 Mar 2024 06:49
Export record
Contributors
Author:
Sandor Szedmak
Author:
John Shawe-Taylor
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics