Shijie Zhong

Directional Kernel Mean Difference: A Fast Signed Statistic for Univariate Distribution Comparison

Shijie Zhong, Jiangfeng Fu

We introduce the Directional Kernel Mean Difference (DKMD), a signed statistic for univariate distribution comparison that preserves the direction of distributional shifts. Unlike the squared Maximum Mean Discrepancy (MMD), which discards directional information by squaring the R…Read more