Multi-Objective Parameter Selection for Classifiers

作者：Muessel Christoph; Lausser Ludwig; Maucher Markus; Kestler Hans A^*

来源：Journal of Statistical Software, 2012, 46(5): 1-27.

• classification
• parameter tuning
• multi-objective optimization
• R

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 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.

• 出版日期2012-1