A Memetic Algorithm Guided by Quicksort for the Error-Correcting Graph Isomorphism Problem

Abstract

Sorting algorithms define paths in the search space of n! permutations based on the information provided by a comparison predicate. We guide a Memetic Algorithm with a new mutation operator. Our mutation operator performs local search following the path traced by the Quicksort mechanism. The comparison predicate and the evaluation function are made to correspond and guide the evolutionary search. Our approach improves previous results for a benchmark of experiments of the Error-Correcting Graph Isomorphism. For this case study, our new Memetic Algorithm achieves a better quality vs effort trade-off and remains highly effective even when the size of the problem grows.