Title :
Deterministic weight evolution algorithm for solving local minima problem
Author :
Ng, Sin-Chun ; Cheung, Chi-Chung ; Shu-Hung Leung
Author_Institution :
Dept. of Comput. & Math., Inst. of Vocational Educ., Chai Wan, Hong Kong
fDate :
6/8/2000 12:00:00 AM
Abstract :
A new algorithm is proposed to enable local minima to be escaped from by changing the weights of a multi-layer neural network in a deterministic way. Simulation results show that the new algorithm always outperforms other traditional methods in achieving global convergence
Keywords :
backpropagation; convergence; evolutionary computation; feedforward neural nets; deterministic weight evolution algorithm; global convergence; local minima problem; multilayer feedforward neural networks; supervised learning;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20000747
