期刊名称：International Journal of Computer and Information Technology
印刷版ISSN：2279-0764
出版年度：2014
卷号：3
期号：2
页码：223
出版社：International Journal of Computer and Information Technology
摘要：Regularized Maximum Mean Discrepancy (RMMD), our novel measure for kernel-based hypothesis testing, excels at hypothesis tests involving multiple comparisons with power control even when sample sizes are small. We derive asymptotic distributions under the null and alternative hypotheses, and assess power control. Outstanding results are obtained on challenging benchmark datasets.
关键词：kernel-based hypothesis testing; Homogeneity ; testing; Multiple comparisons; Power
