Global Minimum Point Search Under Noisy Observations Using Estimator-Type Variable Hierarchical Structure Learning Automata
Yoshio Mogami, Norio BABA, Masaki MATSUSHITA · Transactions of the Society of Instrument and Control Engineers · 1999
The purpose of this paper is to construct a global optimization algorithm of the unknown multimodal objective function under noisy observations. Our algorithm is based on the learning performance of the variable hierarchical structure learning automata, and, in order to reduce the number of iterations, the estimator-type learning algorithm which is the rapdly converging one is used for the learning algorithm of the automata. The numerical experiment is carried out to verify the efficiency of the proposed algorithm, and, from the results, the proposed global optimization algorithm is useful for finding out a global minimum of the unkonwn multimodal objective function under noisy observations.