DocumentCode
3028013
Title
Learning temporal patterns in recurrent neural networks
Author
Doya, Kenji
Author_Institution
Dept. of Math. Eng. & Inf. Phys., Tokyo Univ., Japan
fYear
1990
fDate
4-7 Nov 1990
Firstpage
170
Lastpage
172
Abstract
General learning algorithms for recurrent neural networks that can be used for both discrete-time and continuous-time models are described. They are based on the notion of the derivatives of mappings between functions of time. Simulation results for learning rhythmical sequences are shown. The method can also be applied to networks of higher-order neuron models and multiple time delay connections
Keywords
learning systems; neural nets; continuous-time models; discrete-time; higher-order neuron models; learning systems; multiple time delay connections; recurrent neural networks; rhythmical sequences; temporal patterns; Computer networks; Equations; Humans; Information processing; Intelligent networks; Joining processes; Output feedback; Physics; Recurrent neural networks; Supervised learning;
fLanguage
English
Publisher
ieee
Conference_Titel
Systems, Man and Cybernetics, 1990. Conference Proceedings., IEEE International Conference on
Conference_Location
Los Angeles, CA
Print_ISBN
0-87942-597-0
Type
conf
DOI
10.1109/ICSMC.1990.142085
Filename
142085
Link To Document