Presentation 2002/6/21
Behavior control of an autonomous agent using Flexibly Connected Neural Network
Kazuko TAZAWA, Tomoharu NAGAO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In general, the structure of neural network, such as the layered type or the mutually connected type, etc., is chosen as to the problem. And weights and thresholds of neural network are usually adjusted by the learning method corresponding to the structure. Some given problems, however, cannot be always solved by the neural network of the structure decided beforehand. And it is also difficult for the neural network of the basic structure to process complex information like our brains. Here we present Flexibly Connected Neural Network as a method of constructing arbitrary neural network with optimized the structure and parameters to solve unknown problems. In order to examine the effectiveness, we applied the proposed method to behavior control of an autonomous agent with perceptual aliasing, and analyzed the acquired neural network.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) neural network / genetic algorithm / autonomous agent / behavior control
Paper # PRMU2002-43
Date of Issue

Conference Information
Committee PRMU
Conference Date 2002/6/21(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Behavior control of an autonomous agent using Flexibly Connected Neural Network
Sub Title (in English)
Keyword(1) neural network
Keyword(2) genetic algorithm
Keyword(3) autonomous agent
Keyword(4) behavior control
1st Author's Name Kazuko TAZAWA
1st Author's Affiliation Graduate School of Environment and Information Sciences, Yokohama National University()
2nd Author's Name Tomoharu NAGAO
2nd Author's Affiliation Graduate School of Environment and Information Sciences, Yokohama National University
Date 2002/6/21
Paper # PRMU2002-43
Volume (vol) vol.102
Number (no) 156
Page pp.pp.-
#Pages 8
Date of Issue