Presentation | 2013-09-02 Topics on the Cost in Machine Learning Shotaro AKAHO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The goal of most machine learning algorithms is to minimize a cost function, and thus the cost is a subject of major interest in machine learning research. Bayesian decision theory, which gives a fundamental framework of cost, assumes that the probabilities and cost values are known. However, they are unknown in practical situations, so machine learning techniques are necessary to resolve such uncertainty. In this talk, several interesting research topics on the cost are presented, including relation to privacy preservation and fairness-aware learning that have attracted much attention recently. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Cost-sensitive learning / robustness / Bayesian decision theory / ROC analysis / inverse reinforcement learning |
Paper # | PRMU2013-40,IBISML2013-20 |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 2013/8/26(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 | Pattern Recognition and Media Understanding (PRMU) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Topics on the Cost in Machine Learning |
Sub Title (in English) | |
Keyword(1) | Cost-sensitive learning |
Keyword(2) | robustness |
Keyword(3) | Bayesian decision theory |
Keyword(4) | ROC analysis |
Keyword(5) | inverse reinforcement learning |
1st Author's Name | Shotaro AKAHO |
1st Author's Affiliation | The National Institute of Advanced Industrial Science and Technology (AIST)() |
Date | 2013-09-02 |
Paper # | PRMU2013-40,IBISML2013-20 |
Volume (vol) | vol.113 |
Number (no) | 196 |
Page | pp.pp.- |
#Pages | 2 |
Date of Issue |