CMAC: an associative neural network alternative to backpropagation

The CMAC (cerebellar model arithmetic computer) neural network, an alternative to backpropagated multilayer networks, is described. The following advantages of CMAC are discussed: local generalization, rapid algorithmic computation based on LMS (least-mean-square) training, incremental training, functional representation, output superposition, and a fast practical hardware realization. A geometrical explanation of how CMAC works is provided, and applications in robot control, pattern recognition, and signal processing are briefly described. Possible disadvantages of CMAC are that it does not have global generalization and that it can have noise due to hash coding. Care must be exercised (as with all neural networks) to assure that a low error solution will be learned. >

[1]  J. Albus.  A Theory of Cerebellar Function , 1971 .

[2]  James S. Albus,et al.  Data Storage in the Cerebellar Model Articulation Controller (CMAC) , 1975 .

[3]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[4]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[5]  Enis Ersü,et al.  A New Concept for Learning Control Inspired by Brain Theory , 1984 .

[6]  W. Thomas Miller,et al.  Sensor-based control of robotic manipulators using a general learning algorithm , 1987, IEEE J. Robotics Autom..

[7]  Filson H. Glanz,et al.  Application of a General Learning Algorithm to the Control of Robotic Manipulators , 1987 .

[8]  W. Thomas Miller,et al.  A Nonlinear Learning Controller for Robotic Manipulators , 1987, Other Conferences.

[9]  Enis Ersü,et al.  Hierarchical Learning Control - An Approach with Neuron-Like Associative Memories , 1987, NIPS.

[10]  Bernard Widrow,et al.  Adaptive switching circuits , 1988 .

[11]  W. T. Miller,et al.  Shape Recognition Using A CMAC Based Learning System , 1988, Other Conferences.

[12]  W. Thomas Miller,et al.  Deconvolution using a CMAC neural network , 1988, Neural Networks.

[13]  W. T. Miller,et al.  Real time experiments in neural network based learning control during high speed nonrepetitive robotic operations , 1988, Proceedings IEEE International Symposium on Intelligent Control 1988.

[14]  L. Gordon Kraft,et al.  A Comparison of CMAC Neural Network and Traditional Adaptive Control Systems , 1989, 1989 American Control Conference.

[15]  W. T. Miller,et al.  Deconvolution and nonlinear inverse filtering using a neural network , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[16]  W. Thomas Miller,et al.  Pattern Recognition Using A CMAC Based Learning System , 1989, Other Conferences.

[17]  W. Thomas Miller,et al.  Real-time application of neural networks for sensor-based control of robots with vision , 1989, IEEE Trans. Syst. Man Cybern..

[18]  L. G. Kraft,et al.  Comparison of convergence properties of CMAC neural networks and traditional adaptive controllers , 1989, Proceedings of the 28th IEEE Conference on Decision and Control,.

[19]  W. Thomas Miller,et al.  Real-time dynamic control of an industrial manipulator using a neural network-based learning controller , 1990, IEEE Trans. Robotics Autom..