Automatic Design for Matheuristics

Matheuristics have become widespread and effective methods for tackling the generalized assignment problem (GAP) and many other NP-hard problems. In fact, in this book we have many such methods, ranging from metaheuristics and mathematical programming techniques but mainly to real matheuristics. In these methods we will see no parameter settings, but it is true: all these methods in the end rely on a good parameter setting. Here, in this chapter, we will learn of what can be seen as a good parameter setting. After having seen the parameters and the parameter configuration problem, we turn to the automatic algorithm configuration and then learn about three methods: ParamILS, SMAC, and irace. Then we learn about what these methods are good for and go into some details of automated configuration of existing algorithms, the integration of an algorithm engineering process up to the design of, in part, completely new algorithms. The chapter ends with a little example of how irace can be used to automatically configure a simple iterated local search method for the GAP.