Presentation 1997/5/23
Regularizing the Cascade Correlation Algorithm to Avoid Weight-illgrowth
Qun XU, Kenji NAKAYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper investigates some possible problems of Cascade Correlation algorithm, one of which is the zigzag output mapping caused by weight-illgrowth of the adding hidden unit. Without doubt, it could lead to deteriorate the generialization, especially for regression problems. To solve this problem, we combine Cascade Correlation algorithm with regular-ization theory. In addition, some new regularization terms are proposed in light of special cascade structure. Simulation has shown that regularization indeedly smooth the zigzag output, so that the generialization is improved, especially for functional approximation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) neural network / dynamic learning algorithm / cascade correlation / generialization / regularization
Paper # NC97-3
Date of Issue

Conference Information
Committee NC
Conference Date 1997/5/23(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 ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Regularizing the Cascade Correlation Algorithm to Avoid Weight-illgrowth
Sub Title (in English)
Keyword(1) neural network
Keyword(2) dynamic learning algorithm
Keyword(3) cascade correlation
Keyword(4) generialization
Keyword(5) regularization
1st Author's Name Qun XU
1st Author's Affiliation Fac. Engineering, Kanazawa Univ.()
2nd Author's Name Kenji NAKAYAMA
2nd Author's Affiliation Fac. Engineering, Kanazawa Univ.
Date 1997/5/23
Paper # NC97-3
Volume (vol) vol.97
Number (no) 69
Page pp.pp.-
#Pages 8
Date of Issue