Presentation 1999/10/21
Support Vector Machine and Restoration Problem
Koji Tsuda,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Restoration is the task to obtain the original signal from sampled points of the degraded signal. When the class labels, which are assigned by a teacher, are regarded as the degradation of discriminant function, the learning of classifier can be formulated as restoration. We will show that the support vector learning can be formulated as restoration based on noiseless model. Based on this analysis, we will propose a SV learning based on noisy model. In experiments, noisy SVM is compared with the conventional one with respect to generalization ability.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Pattern recognition / Support Vector Machine / MAP estimate / Wiener Filter
Paper # NC99-43
Date of Issue

Conference Information
Committee NC
Conference Date 1999/10/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Support Vector Machine and Restoration Problem
Sub Title (in English)
Keyword(1) Pattern recognition
Keyword(2) Support Vector Machine
Keyword(3) MAP estimate
Keyword(4) Wiener Filter
1st Author's Name Koji Tsuda
1st Author's Affiliation Machine Understanding Division, Electrotechnical Laboratory()
Date 1999/10/21
Paper # NC99-43
Volume (vol) vol.99
Number (no) 382
Page pp.pp.-
#Pages 8
Date of Issue