Abstract

A method for causing a neural circuit model to learn typical past control results of a process and using the neural circuit model for supporting an operation of the process. The neural circuit model is caused to learn by using, as input signals, a typical pattern of values of input variables at different points in time and, as a teacher signal, its corresponding values of the control variable. An unlearned pattern of input variables is inputted to the thus-learned neuron circuit model, whereby a corresponding value of the control variable is determined. Preferably, plural patterns at given time intervals can be simultaneously used as patterns to be learned.

Other References

• Wasserman et al, "Neural Network, Part 2", IEEE Expert, 1988
• Widrow et al, "Neural Nets for Adaptive Filtering and Adaptive Pattern Recognition," IEEE 1988
• Waibel et al, "Phoneme Recognition: Neural Networks vs. Hidden Markov Models," IEEE ICASSP Conf., 1988
• Macukow et al, "Neural Network Model using a normalized inner product as measure of similarity", IEEE ICNN, 1988
• Hecht-Nielsen, "Applications of counterpropagation Networks," Neural Networks, vol. 1, pp. 131-139, 1988
• Beynon et al, "The Implementation of Multi-Layer Perceptrons on Transputer Networks," PPTBM, Proceedings of the 7th OCCAM User, 1987
• Nagata, et al, "Neurocomputer and its Application to Robot Control", Fujitsu, Jun. 1988, vol. 39, No. 3, pp. 175-184
• Control Engineering, Feb. 1975, New York, U.S., pp. 50-53, Can A Process "Train" its Control System? by S.J. Bailey
• Biological Cybernetics, "A herarchical Neural-Network Model for Control and Learning of Voluntary Movement", M. Kawato, et al., vol. 57, No. 3, Oct. 1987
• Advances in Instrumentation, "Applications of Artificial Neural Systems in Robotic Welding", by B.A. Cleveland, vol. 43, No. PART 04, Jan. 1, 1988
• "Suidokyokai Zasshi (Journal of the Water Service Workers' Association)", No. 431, p. 28, Aug. 1970
• Learning Internal Representations in the Coulomb Energy Network, Scofield, Jul. 1988, pp. 271-276
• "A Class of Gradient-Estimating Algorithms for Reinforcement Learning in Neural Networks" IEEE 1st International Conference on Neural Networks by Ronald J. Williams, Jun. 1987
• "A Learning Algorithm for Analog Full recurrent Neural Networks" by Michael Gherrity, Jun. 18, 1980
• "Training Time-Dependence in Neural Networks", IEEE 1st International Conference on Neural Networks by Richard Rohwer, Jun. 1987
• "Neural Networks Primer" AI Expert, Feb. 1989
• Layered Neural Nets for Pattern Recognition by Widrow, et al., Speech and Signal Processing, vol. 36, No. 7, Jul. 1988
• Parallel Distributed Processing, "Learning Internal Representations by Error Propagation", D.E. Rumelhart, vol. 1: Foundations, 1986, The MIT Pres