Presentation 1998/6/18
Graphic Transformation Method applied to Asymmetric Neural Networks
Kenichiro Mogi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) King and Altman's graphic method describes a way of expressing the steady state distribution in a system of discrete states. The graphic transformation method is developed on top of the King & Altman method. In the graphic transformation method, we consider how each terms in the distribution weight of the system is transformed as a result of the graphic transformation. We can then get a separate expression for the contribution of the asymmetric synaptic weights in a neural network. It is found that in an asymmetric neural network it is not possible to assign the energy value locally to a specific neural state.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) asymmetric connection / graph theory / graphic transformation method / neural network / steady state
Paper # NC98-23,HIP98-14
Date of Issue

Conference Information
Committee NC
Conference Date 1998/6/18(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Vice Chair

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Graphic Transformation Method applied to Asymmetric Neural Networks
Sub Title (in English)
Keyword(1) asymmetric connection
Keyword(2) graph theory
Keyword(3) graphic transformation method
Keyword(4) neural network
Keyword(5) steady state
1st Author's Name Kenichiro Mogi
1st Author's Affiliation Sony Computer Science Laboratory()
Date 1998/6/18
Paper # NC98-23,HIP98-14
Volume (vol) vol.98
Number (no) 128
Page pp.pp.-
#Pages 8
Date of Issue