Multi-Objective Parameter Selection for Classifiers

Christoph Müssel, Ludwig Lausser, Markus Maucher, Hans A. Kestler
2012 Journal of Statistical Software  
Setting the free parameters of classifiers to different values can have a profound impact on their performance. For some methods, specialized tuning algorithms have been developed. These approaches mostly tune parameters according to a single criterion, such as the cross-validation error. However, it is sometimes desirable to obtain parameter values that optimize several concurrent -often conflicting -criteria. The TunePareto package provides a general and highly customizable framework to
more » ... optimal parameters for classifiers according to multiple objectives. Several strategies for sampling and optimizing parameters are supplied. The algorithm determines a set of Pareto-optimal parameter configurations and leaves the ultimate decision on the weighting of objectives to the researcher. Decision support is provided by novel visualization techniques.
doi:10.18637/jss.v046.i05 fatcat:mouyb6hvwnbnhetn7r75mhlpju