Abstract

The irace package implements a flexible tool for the automatic configuration of algorithms. However, irace itself has specific parameters to customize the search process according to the tuning scenario. In this paper, we analyze five parameters of irace: the number of iterations, the number of instances seen before the first elimination test, the maximum number of elite configurations, the statistical test and the confidence level of the statistical test. These parameters define some key aspects of the way irace identifies good configurations. Originally, their values have been set based on rules of thumb and an intuitive understanding of the configuration process. This work aims at giving insights about the sensitivity of irace to these parameters in order to guide their setting and further improvement of irace.