Presentation | 2005-02-24 The Theory and Algorithm of Semi-supervised Learning Naonori UEDA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In classifier design, labeled data are often fairly expensive to acquire beacase the class labels are identified by human experts. Semi-supervised learning provides methods for effectively using a large amount of unlabeled data with a small amount of labeled data when trainig a classifier. Semi-supervised learning cosists of parametric approch based on probabilistic model and non-parametric approach in which spectral method is used based on data similarities. This report gives a review on theory and algorithm of both approaches. In particular, for non-parametric approach. the importance of metric learning is shown and a method that simultaneously performs posterior propagation and metric learning is newly presented with preliminary result using artificial data. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | semi-supervised learning / pattern recognition / non-parametric approach / spectral clustering / metric learning |
Paper # | NLC2004-101,PRMU2004-183 |
Date of Issue |
Conference Information | |
Committee | NLC |
---|---|
Conference Date | 2005/2/17(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 | Natural Language Understanding and Models of Communication (NLC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | The Theory and Algorithm of Semi-supervised Learning |
Sub Title (in English) | |
Keyword(1) | semi-supervised learning |
Keyword(2) | pattern recognition |
Keyword(3) | non-parametric approach |
Keyword(4) | spectral clustering |
Keyword(5) | metric learning |
1st Author's Name | Naonori UEDA |
1st Author's Affiliation | NTT Communication Science Laboratories, NTT Corporation() |
Date | 2005-02-24 |
Paper # | NLC2004-101,PRMU2004-183 |
Volume (vol) | vol.104 |
Number (no) | 667 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |