Cellular SRN Trained by Extended Kalman Filter Shows Promise for ADP

Cellular simultaneous recurrent neural network has been suggested to be a function approximator more powerful than the MLP's, in particular for solving approximate dynamic programming problems. The 2D maze navigation has been considered as a proof-of-concept task. Present work improves the prev...

Full description

Saved in:
Bibliographic Details
Published in:The 2006 IEEE International Joint Conference on Neural Network Proceedings pp. 506 - 510
Main Authors: Ilin, R., Kozma, R., Werbos, P.J.
Format: Conference Proceeding
Language:English
Published: IEEE 2006
Subjects:
Cellular networks
Convergence
Cost function
Dynamic programming
Electronic mail
Equations
Learning
Navigation
Neural networks
Recurrent neural networks
ISBN:9780780394902, 0780394909
ISSN:2161-4393
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cellular simultaneous recurrent neural network has been suggested to be a function approximator more powerful than the MLP's, in particular for solving approximate dynamic programming problems. The 2D maze navigation has been considered as a proof-of-concept task. Present work improves the previous results by training the network with extended Kalman filter (EKF). The original EKF algorithm has been slightly modified. The speed of convergence has been improved by several orders of magnitude in comparison with the earlier results. The implications of this improvement are discussed.
ISBN:9780780394902
0780394909
ISSN:2161-4393
DOI:10.1109/IJCNN.2006.246724