Title :
The autoregressive backpropagation algorithm
Author :
Leighton, Russell R. ; Conrath, Bartley C.
Author_Institution :
Mitre Corp., McLean, VA, USA
Abstract :
Describes an extension to error backpropagation that allows the nodes in a neural network to encode state information in an autoregressive `memory´. This neural model gives such networks the ability to learn to recognize sequences and context-sensitive patterns. Building upon the work of A. Wieland (1990) concerning nodes with a single feedback connection, the authors generalize the method to n feedback connections and address stability issues. The learning algorithm is derived, and a few applications are presented
Keywords :
feedback; learning systems; neural nets; pattern recognition; stability; autoregressive backpropagation algorithm; context-sensitive patterns; feedback connections; learning algorithm; neural network; sequence recognition; stability; state information encoding; Backpropagation algorithms; Context modeling; Delay; Difference equations; Digital filters; Neural networks; Neurofeedback; Neurons; Output feedback; Stability;
Conference_Titel :
Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-0164-1
DOI :
10.1109/IJCNN.1991.155362
