Title :
Conditioned adaptive behavior from Kalman filter trained recurrent networks
Author :
Feldkamp, Lee A. ; Prokhorov, Danil V. ; Feldkamp, Timothy M.
Author_Institution :
Res. & Adv. Eng., Ford Motor Co., Dearborn, MI, USA
Abstract :
We demonstrate that a fixed-weight neural network can be trained with Kalman filter methods to exhibit input-output behavior that depends on which of two conditioning tasks had been performed a substantial number of time steps in the past. This behavior can also be made to survive an intervening interference task.
Keywords :
Kalman filters; learning (artificial intelligence); recurrent neural nets; Kalman filter trained recurrent networks; conditioned adaptive behavior; fixed-weight neural network; input output behavior; intervening interference task; Adaptive systems; Chaos; Control systems; Interference; Neural networks; Recurrent neural networks;
Conference_Titel :
Neural Networks, 2003. Proceedings of the International Joint Conference on
Print_ISBN :
0-7803-7898-9
DOI :
10.1109/IJCNN.2003.1224052
