Title :
Two dimensional function learning using CMAC neural network with optimized weight smoothing
Author :
Pallotta, Jeremy ; Kraft, L.G.
Author_Institution :
Dept. of Electr. & Comput. Eng., New Hampshire Univ., Durham, NH, USA
Abstract :
This paper compares the traditional CMAC neural network weight update algorithm with a new optimized weight smoothing approach. Although CMAC learns functions rapidly, there is an inherent “roughness” to the approximation caused by spikes in the weight space even when the function being learned is relatively smooth. The new CMAC weight smoothing update scheme produces better approximations for a large class of functions
Keywords :
cerebellar model arithmetic computers; learning (artificial intelligence); optimisation; 2D function learning; CMAC neural network weight update algorithm; CMAC weight smoothing update scheme; optimized weight smoothing; rough approximation; weight space spikes; Biomembranes; Electronic mail; Equations; Function approximation; Neural networks; Process control; Signal processing; Signal processing algorithms; Smoothing methods; Vibration control;
Conference_Titel :
American Control Conference, 1999. Proceedings of the 1999
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-4990-3
DOI :
10.1109/ACC.1999.782804
