Presentation | 2010-01-21 Superfast Probabilistic Classifier Masashi SUGIYAMA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Kernel logistic regression (KLR) is a powerful and flexible classification algorithm, which possesses an ability to provide the confidence of class prediction. However, its training-typically carried out by Newton's method or quasi-Newton methods-is rather time-consuming. In this paper, we propose an alternative probabilistic classification algorithm called Least-Squares Probabilistic Classifier (LSPC). The solution of LSPC can be computed analytically just by solving a system of linear equations, so LSPC is computationally efficient and stable. Through experiments, we show that the computation time of LSPC is faster than that of KLR by the factor 100 with comparable accuracy. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | probabilistic classification / kernel logistic regression / class-posterior probability / squared-loss |
Paper # | CQ2009-74,PRMU2009-173,SP2009-114,MVE2009-96 |
Date of Issue |
Conference Information | |
Committee | MVE |
---|---|
Conference Date | 2010/1/14(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 | Media Experience and Virtual Environment (MVE) |
---|---|
Language | ENG |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Superfast Probabilistic Classifier |
Sub Title (in English) | |
Keyword(1) | probabilistic classification |
Keyword(2) | kernel logistic regression |
Keyword(3) | class-posterior probability |
Keyword(4) | squared-loss |
1st Author's Name | Masashi SUGIYAMA |
1st Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology:JST() |
Date | 2010-01-21 |
Paper # | CQ2009-74,PRMU2009-173,SP2009-114,MVE2009-96 |
Volume (vol) | vol.109 |
Number (no) | 376 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |