Presentation	1996/2/3 Analysis of Improvement Effect for Generalization Error of Ensemble Estimators Naonori UEDA, Ryohei NAKANO,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	It has been empirically shown that a better estimate with less generalization error can be obtained by averaging outputs of multiple estimators. However, so far the theoretical study of how the averaging reduces the generalization error has been superficial. This report presents an analytical result for the generalization error of ensemble regression estimators. That is, we derive a general expression of the ensemble generalization error by using factors of interest (bias, variance, covariance, and noise variance) and show how the generalization error is affected by each of them. The result of a simulation is shown to verify our analytical result. A practically important problem, the sample size effect on the ensemble approach, is also discussed.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	Generalization error / Ensemble learning / Nonlinear regression
Paper #	NC95-115
Date of Issue

Conference Information
Committee	NC
Conference Date	1996/2/3(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)	Analysis of Improvement Effect for Generalization Error of Ensemble Estimators
Sub Title (in English)
Keyword(1)	Generalization error
Keyword(2)	Ensemble learning
Keyword(3)	Nonlinear regression
1st Author's Name	Naonori UEDA
1st Author's Affiliation	NTT Communication Science Labor atories()
2nd Author's Name	Ryohei NAKANO
2nd Author's Affiliation	NTT Communication Science Labor atories
Date	1996/2/3
Paper #	NC95-115
Volume (vol)	vol.95
Number (no)	506
Page	pp.pp.-
#Pages	8
Date of Issue