John Holland, Melanie Mitchell
Paper #: 93-06-037
In this paper we review some previously published experimental results in which a simple hill-climbing algorithm--Random Mutation Hill-Climbing (RMHC)--significantly outperforms a genetic algorithm on a simple “Royal Road” function. We present an analysis of RMHC followed by an analysis of an “idealized” genetic algorithm (IGA) that is, in turn, significantly faster than RMHC. We isolate the features of the IGA that allow for this speedup, and discuss how these features can be incorparated into a real GA and a fitness landscape, making the GA better approximate the IGA. We use these features to design a modified version of the previously published experiments, and give new experimental results comparing the GA and RMHC.