%0 Journal Article
%T Multi-Objective Parameter Selection for Classifers
%A Christoph Mussel
%A Ludwig Lausser
%A Markus Maucher
%A Hans A. Kestler
%J Journal of Statistical Software
%D 2012
%I University of California, Los Angeles
%X 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 select optimal parameters for classifiers according to multiple objectives. Several strategies for sampling andoptimizing parameters are supplied. The algorithm determines a set of Pareto-optimal parameter configuration and leaves the ultimate decision on the weighting of objectives to the researcher. Decision support is provided by novel visualization techniques.
%K classication
%K parameter tuning
%K multi-objective optimization
%K R
%U http://www.jstatsoft.org/v46/i05/paper