Presentation 2012-11-08
Efficient SVM Bootstrap Computation by Parametric Programming
Yoshiki SUZUKI, Kohei OGAWA, Ichiro TAKEUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we study statistical variability estimation of the support vector machine (SVM) by using bootstrap method. Training a large number of the SVMs for each of the thousands of bootstrap samples is quite time-consuming. We therefore develop a new algorithm for efficiently computing these large set of the SVMs by introducing two techniques called parametric programming and sphere test. Parametric programming allows us to efficiently compute the optimal solutions of two similar problems, while sphere test enables us to estimate the similarity or distance between two different SVMs. We demonstrate the effectiveness of our algorithm through numerical experiments on real data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Support Vector Machine / Bootstrap / Parametric Programing / Sphere Test
Paper # IBISML2012-73
Date of Issue

Conference Information
Committee IBISML
Conference Date 2012/10/31(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 Information-Based Induction Sciences and Machine Learning (IBISML)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Efficient SVM Bootstrap Computation by Parametric Programming
Sub Title (in English)
Keyword(1) Support Vector Machine
Keyword(2) Bootstrap
Keyword(3) Parametric Programing
Keyword(4) Sphere Test
1st Author's Name Yoshiki SUZUKI
1st Author's Affiliation Department of Engineering, Nagoya Institute of Technology()
2nd Author's Name Kohei OGAWA
2nd Author's Affiliation Department of Engineering, Nagoya Institute of Technology
3rd Author's Name Ichiro TAKEUCHI
3rd Author's Affiliation Department of Engineering, Nagoya Institute of Technology
Date 2012-11-08
Paper # IBISML2012-73
Volume (vol) vol.112
Number (no) 279
Page pp.pp.-
#Pages 6
Date of Issue