Training recurrent networks using the extended Kalman filter

The author describes some relationships between the extended Kalman filter (EKF) as applied to recurrent net learning and some simpler techniques that are more widely used. In particular, making certain simplifications to the EKF gives rise to an algorithm essentially identical to the real-time recurrent learning (RTRL) algorithm. Since the EKF involves adjusting unit activity in the network, it also provides a principled generalization of the teacher forcing technique. Preliminary simulation experiments on simple finite-state Boolean tasks indicated that the EKF can provide substantial speed-up in number of time steps required for training on such problems when compared with simpler online gradient algorithms. The computational requirements of the EKF are steep, but scale with network size at the same rate as RTRL.<<ETX>>

[1]  L. Mcbride,et al.  Optimization of time-varying systems , 1965 .

[2]  A.H. Haddad,et al.  Applied optimal estimation , 1976, Proceedings of the IEEE.

[3]  S. Thomas Alexander,et al.  Adaptive Signal Processing , 1986, Texts and Monographs in Computer Science.

[4]  Sharad Singhal,et al.  Training Multilayer Perceptrons with the Extende Kalman Algorithm , 1988, NIPS.

[5]  Ronald J. Williams,et al.  Experimental Analysis of the Real-time Recurrent Learning Algorithm , 1989 .

[6]  Richard Rohwer,et al.  The "Moving Targets" Training Algorithm , 1989, NIPS.

[7]  James L. McClelland,et al.  Finite State Automata and Simple Recurrent Networks , 1989, Neural Computation.

[8]  David Zipser,et al.  A Subgrouping Strategy that Reduces Complexity and Speeds Up Learning in Recurrent Networks , 1989, Neural Computation.

[9]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[10]  Jing Peng,et al.  An Efficient Gradient-Based Algorithm for On-Line Training of Recurrent Network Trajectories , 1990, Neural Computation.

[11]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[12]  Ronald J. Williams,et al.  Gradient-based learning algorithms for recurrent connectionist networks , 1990 .

[13]  Lee A. Feldkamp,et al.  Decoupled extended Kalman filter training of feedforward layered networks , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.