Presentation 2003/7/21
Learning the Data Region using Extreme-value Statistics
Kazuho WATANABE, Sumio WATANABE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the field of pattern recognition or outlier detection, it is desirable to estimate the region where the data of a particular class are generated. In other words, precise prediction is realized by accurately estimating the support of the distribution that generates the data. Considering the 1-dimensional distribution whose support is a finite interval, the data region is characterized by maximum value and the minimum value in the samples. Limiting distributions of thses values have been studied in the extreme-value theory in statistics. In this research, we propose a method to estimate the data region using the maximum value and the minimum value in the samples. We calculate the average loss of the estimator, and derive the optimal estimators for given loss functions.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) data region / limiting distribution / extreme value statistics
Paper # NC2003-30
Date of Issue

Conference Information
Committee NC
Conference Date 2003/7/21(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 JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Learning the Data Region using Extreme-value Statistics
Sub Title (in English)
Keyword(1) data region
Keyword(2) limiting distribution
Keyword(3) extreme value statistics
1st Author's Name Kazuho WATANABE
1st Author's Affiliation Depart of Advanced Applied Electronics, Tokyo Institute of Technology()
2nd Author's Name Sumio WATANABE
2nd Author's Affiliation P&I Lab, Tokyo Institute of Technology
Date 2003/7/21
Paper # NC2003-30
Volume (vol) vol.103
Number (no) 227
Page pp.pp.-
#Pages 6
Date of Issue