Evolution Strategies Learned with Automatic Termination Criteria

Abstract

Although several attempts have been made to modify the original versions of Evolutionary Algorithms (EAs), they are not learned with automatic termination criteria. In general, EAs cannot decide when or where they can terminate. In this paper, we modify Evolution Strategies (ESs) with new termination criteria. The proposed method is called ESs Learned with Automatic Termination (ESLAT). In the ESLAT method, a so-called Gene Matrix (GM) is constructed to equip the search process with a self-check to judge how much exploration has been done. Moreover, an especial mutation operation called “Mutagenesis” is defined to achieve more efficient and faster exploration process. The computational experiments show the efficiency of the ESLAT method. Keywords—Evolution Strategies, Evolutionary Algorithms, Heuristics, Global Optimization, Termination Criteria, Mutagenesis