Presentation 2002/12/6
Probabilistic State Transitions in Spiking Neural Network
Akiko IWAIZUMI, Ryoko FUTAMI, Nozomu HOSHIMIYA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Male Bengalese finches sing in a fashion that song elements array stochastically each other, which means transition in song elements can be represented as a probabilistic transition matrix. There is no model found that achieve probabilistic state transitions using spiking neuron model. We adopt Hodgkin-Huxley neuron model and propose the model that introduces the network X with auto-associative memory function and mutually inhibited neuron chains with subthreshold noise to make a state of X transit stochastically. We analysed firing rates and temporal structures of each stair of neuron chains to the connection value between X and chains and the number of chains. Numerical simulation revealed that the state transition depends on the firing rate of the last neuron of chain and the connection values determine the firing rate.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Spiking neuron / probabilistic transition matrix / neuron chain / mutual inhibition subthreshold noise
Paper # NC2002-97
Date of Issue

Conference Information
Committee NC
Conference Date 2002/12/6(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) Probabilistic State Transitions in Spiking Neural Network
Sub Title (in English)
Keyword(1) Spiking neuron
Keyword(2) probabilistic transition matrix
Keyword(3) neuron chain
Keyword(4) mutual inhibition subthreshold noise
1st Author's Name Akiko IWAIZUMI
1st Author's Affiliation Department of Electronic Engineering, Tohoku University()
2nd Author's Name Ryoko FUTAMI
2nd Author's Affiliation Department of Electronic Engineering, Tohoku University
3rd Author's Name Nozomu HOSHIMIYA
3rd Author's Affiliation Department of Electronic Engineering, Tohoku University
Date 2002/12/6
Paper # NC2002-97
Volume (vol) vol.102
Number (no) 508
Page pp.pp.-
#Pages 6
Date of Issue