Presentation | 1998/3/19 Time-domain behavior stabilization of complex-valued recurrent neural networks using relative-minimization learning Akira Hirose, Hirofumi Onishi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Relative-minimization learning using additional random teacher signals is newly proposed for time-domain behavior stabilization of recurrent neural networks. Although the recurrent neural networks can deal with time-sequential data, they tend to have an unstable behavior showing positive Lyapunov exponents. Shortly speaking, the proposed method superimposes a type of basin upon a dynamics-determining hypersurface in an information vector field. This process is equivalent to the relative minimization of the error function in the input-signal partial space. Experiments demonstrate that the relative-minimization learning suppresses positive values of Lyapunov exponents down to zero or negative, which means that the behavior in time-domain is successfully stabilized. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Recurrent neural network / Time-domain stability / Lyapunov spectrum |
Paper # | |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1998/3/19(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) | Time-domain behavior stabilization of complex-valued recurrent neural networks using relative-minimization learning |
Sub Title (in English) | |
Keyword(1) | Recurrent neural network |
Keyword(2) | Time-domain stability |
Keyword(3) | Lyapunov spectrum |
1st Author's Name | Akira Hirose |
1st Author's Affiliation | Research Center for Advanced Science and Technology(RCAST), University of Tokyo() |
2nd Author's Name | Hirofumi Onishi |
2nd Author's Affiliation | Presently with the NTT visual Communication Sector |
Date | 1998/3/19 |
Paper # | |
Volume (vol) | vol.97 |
Number (no) | 623 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |