Presentation | 2005/6/16 Independent component separation by pattern representation in two random symmetric networks connected via anti-Hebbian synapses Koji WADA, Koji KURATA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper, we propose a new self-organnizing model and the learning rule of the model. The model consisits of two random symmetric network connected with anti-Hebbian synapses. We confirmed by computational experiment that when input signals that have two independent components are given to the model, each independent components are extracted into two random symmetric network of the model. Conventional self-organizing maps have Mexican-hat type intra-layer connections. So, they extract the information by way of relating the input to the winner-neuron. Our model introduces the random symmetric intra-layer connections. So, to extract the information, instead of relating the input to the winner-neuron, we relate the input to the firing pattern. That is, our model is the self-organizing model of pattern representation, that extract the independent component into two random symmetric network. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Anti-Hebbian learning / random symmetric network / pattern representation / Independent Component Analysis / Self-organization |
Paper # | NC2005-19 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2005/6/16(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) | Independent component separation by pattern representation in two random symmetric networks connected via anti-Hebbian synapses |
Sub Title (in English) | |
Keyword(1) | Anti-Hebbian learning |
Keyword(2) | random symmetric network |
Keyword(3) | pattern representation |
Keyword(4) | Independent Component Analysis |
Keyword(5) | Self-organization |
1st Author's Name | Koji WADA |
1st Author's Affiliation | Dept. of Electrical Engineering, Kochi National College of Technology() |
2nd Author's Name | Koji KURATA |
2nd Author's Affiliation | Faculty of Engineering, Ryukyu University |
Date | 2005/6/16 |
Paper # | NC2005-19 |
Volume (vol) | vol.105 |
Number (no) | 130 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |