Title :
Neural network for global optimization
Author :
Cheung, Kwok-Wai ; Lee, Tong
Author_Institution :
Dept. of Electron. Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong
Abstract :
A tunneling neural network that can be used in conjunction with a Hopfield network to solve a global minimization problem is proposed. When the two are used together, the global minimum of the corresponding Hopfield energy can be attained. Simulation was performed for the traveling salesman problem with four to ten cities, and a solution very near to the global minimum was reached for all the cases investigated
Keywords :
Hopfield neural nets; minimisation; operations research; Hopfield energy; Hopfield network; global minimization; global optimization; simulation; traveling salesman problem; tunneling neural network; Annealing; Cities and towns; Constraint optimization; Constraint theory; Minimization methods; Multidimensional systems; Neural networks; Performance analysis; Traveling salesman problems; Tunneling;
Conference_Titel :
Neuroinformatics and Neurocomputers, 1992., RNNS/IEEE Symposium on
Conference_Location :
Rostov-on-Don
Print_ISBN :
0-7803-0809-3
DOI :
10.1109/RNNS.1992.268610
