Strictly Proper Kernel Scoring Rules and Divergences with an Application to Kernel Two-Sample Hypothesis Testing
This provides a theoretical framework for kernel-based statistical testing, with incremental improvements for researchers in machine learning and statistics.
The authors developed a general Kernel Scoring rule and associated Kernel Divergence in Reproducing Kernel Hilbert Spaces, showing it includes Maximum Mean Discrepancy as a special case and incorporates more information about projected embedded distributions. They demonstrated improved two-sample hypothesis testing results by integrating information from different Kernel Divergences using a one-class classifier.
We study strictly proper scoring rules in the Reproducing Kernel Hilbert Space. We propose a general Kernel Scoring rule and associated Kernel Divergence. We consider conditions under which the Kernel Score is strictly proper. We then demonstrate that the Kernel Score includes the Maximum Mean Discrepancy as a special case. We also consider the connections between the Kernel Score and the minimum risk of a proper loss function. We show that the Kernel Score incorporates more information pertaining to the projected embedded distributions compared to the Maximum Mean Discrepancy. Finally, we show how to integrate the information provided from different Kernel Divergences, such as the proposed Bhattacharyya Kernel Divergence, using a one-class classifier for improved two-sample hypothesis testing results.