Title :
Threshold net synthesis using simulated annealing
Author_Institution :
Dept. of Comput. Eng., Minnesota Univ., Duluth, MN, USA
Abstract :
The author describes the development of a novel threshold net synthesis algorithm that is based on decomposition by geometrical characteristics of threshold functions. A concept of center of gravity and its related theorems are discussed. A simulated annealing algorithm is adapted for the optimal decomposition of functions with many variables. The significance of the proposed method is not in the exact decomposition of the given function, but in finding the best decomposition with a limited number of threshold elements, from which interesting engineering applications are available. Applications of threshold nets based on the proposed decomposition method are discussed
Keywords :
neural nets; simulated annealing; threshold elements; threshold logic; decomposition; engineering applications; exact decomposition; geometrical characteristics; simulated annealing algorithm; threshold elements; threshold functions; threshold net synthesis algorithm; Arithmetic; Character generation; Computational modeling; Cost function; Gravity; Input variables; Iterative algorithms; Logic; Simulated annealing; Testing;
Conference_Titel :
Southeastcon '91., IEEE Proceedings of
Conference_Location :
Williamsburg, VA
Print_ISBN :
0-7803-0033-5
DOI :
10.1109/SECON.1991.147927
