A Real-Value Parameter Function Optimization Algorithm using Repeated Adaptive Local Search

A simple and easy to implement but very effective algorithm for solving real-value parameter optimization problems is introduced in this paper. The main idea of the perform a local search repeatedly on prospective subregion where optimal solution may be located. randomly samples number solutions given subregion. If new best-so- far has been found, center moved based best-so-far size gradually reduced by predefined shrinking rate. Otherwise, not using This process repeated instances so that focused smaller To enhance likelihood achieving an solution, many rounds are performed. Each round starts with initial space. According experiment results, proposed algorithm, though simple, can outperform some well-known algorithms testing functions.