Revision as of 18:46, 27 March 2020 edit Michael Hardy (talk \| contribs) Administrators 210,597 edits →Distribution regression ← Previous edit		Revision as of 18:52, 27 March 2020 edit undo Michael Hardy (talk \| contribs) Administrators 210,597 edits →Measuring distance between distributions Next edit →
Line 196: === Measuring distance between distributions === The '''maximum mean discrepancy~~'''~~ (MMD)''' is a distance-measure between distributions <math>P(X)</math> and <math>Q(Y)</math> which is defined as the squared distance between their embeddings in the RKHS <ref name = "SongThesis" /> :<math>\text{MMD}(P,Q) = \left \\| \mu_X - \mu_Y \right \\|_{\mathcal{H}}^2 </math>

Kernel embedding of distributions: Difference between revisions