Presentation 2002/6/27
Analysis of 1/ƒ characteristics of Neural Networks with the Fokker-Planck Equation : 1/ƒ characteristics with the Fokker-Planck Equation
Yoshinobu KAMITANI, Ikuo MATSUBA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We have obtained 1/ƒ spectra with the simplifed neural networks, theoretically. And we have built up the neural networks in computer and confirmed 1/ƒ spectra successfully. However, we didn't consider the stability of the solutions that exhibit 1/ƒ spectra. Now, with the Fokker-Planck equation, which deals with the flow of the probability, we discuss the properties of those solutions.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) 1/ƒ spectra / neural network / self-similarity / stability / Fokker-Planck equation
Paper # NLP2002-27
Date of Issue

Conference Information
Committee NLP
Conference Date 2002/6/27(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 Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Analysis of 1/ƒ characteristics of Neural Networks with the Fokker-Planck Equation : 1/ƒ characteristics with the Fokker-Planck Equation
Sub Title (in English)
Keyword(1) 1/ƒ spectra
Keyword(2) neural network
Keyword(3) self-similarity
Keyword(4) stability
Keyword(5) Fokker-Planck equation
1st Author's Name Yoshinobu KAMITANI
1st Author's Affiliation Matsuba Lab., Graduate School of Science and Technology, Chiba University()
2nd Author's Name Ikuo MATSUBA
2nd Author's Affiliation Department of Information and Image Sciences, Faculty of Engineering, Chiba University
Date 2002/6/27
Paper # NLP2002-27
Volume (vol) vol.102
Number (no) 181
Page pp.pp.-
#Pages 5
Date of Issue