Assessing efficiency of different evolutionary strategies playing MasterMind

A MasterMind player must find out a secret combination (set by another player) by playing others of the same kind and using the hints obtained as a response (which reveal how close the played combination is to the secret one) to produce new combinations. Despite having been researched for a number of years, there are still many open issues: finding a strategy to select the next combination to play that is able to consistently obtain good results at any problem size, and also doing so in as little time as possible. In this paper we cast the solution of MasterMind as a constrained optimization problem, introducing a new fitness function for evolutionary algorithms that takes that fact into account, and compare it to other approaches (exhaustive/heuristic and evolutionary), finding that it is able to obtain consistently good solutions, and in as little as 30% less time than previous evolutionary algorithms.