Title :
Intelligent adaptable systems
Author :
Géczy, Peter ; Usui, Shiro ; Amari, Shun-ichi
Author_Institution :
RIKEN Brain Sci. Inst., Saitama, Japan
fDate :
6/24/1905 12:00:00 AM
Abstract :
The next step in neural network field is an integration of several problem domains into a single functioning system. Until now various learning features have been treated in isolation. Fusion of the specific tasks performed during the learning results in a system with novel qualities. The article presents an intelligent adaptable system that integrates sample selection, connection adaptation, structure adaptation and knowledge acquisition. All four subsystems operate dynamically on neural network during the learning
Keywords :
knowledge acquisition; knowledge based systems; learning (artificial intelligence); neural nets; connection adaptation; intelligent adaptable systems; knowledge acquisition; learning features; neural network; sample selection; structure adaptation; Artificial neural networks; Biological neural networks; Information processing; Intelligent structures; Intelligent systems; Isolation technology; Joining processes; Knowledge acquisition; Microstructure; Multidimensional systems;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007780