Self-adaptive Baldwinian search in hybrid genetic algorithms

The problem of proper utilization of the search time to adapt a hybrid to a given problem can be viewed as a problem of finding optimal control parameter settings. The algorithm’s time utilization can be optimized through adapting the local search duration. Evolving this control parameter via genetic operations is one possible way to achieve this adaptation. However, the hindering effect can obstruct the self-adaptive ability of the Baldwinian search. Local search methods with narrow steps and the use of the local search duration to discriminate between solutions can help to alleviate this problem.