Learning automata induced artificial bee colony for noisy optimization

We propose two extensions of the traditional artificial bee colony algorithm to proficiently optimize noisy fitness. The first strategy is referred to as stochastic learning automata induced adaptive sampling. It is employed with an aim to judiciously select the sample size for the periodic fitness evaluation of a trial solution, based on the fitness variance in its local neighborhood. The local neighborhood fitness variance is here used to capture the noise distribution in the local surrounding of a candidate solution of the noisy optimization problem. The second strategy is concerned with determining the effective fitness estimate of a trial solution using the distribution of its noisy fitness samples, instead of direct averaging of the samples. Computer simulations undertaken on the noisy versions of a set of 28 benchmark functions reveal that the proposed algorithm outperforms its contenders with respect to function error value in a statistically significant manner.