It’s All in the Mix: Wasserstein Classification and Regression with Mixed Features

The authors acknowledge the detailed and constructive feedback provided by the department editor, the anonymous associate editor, and the two anonymous referees that has substantially improved the quality of this manuscript. The authors are especially indebted to the reviewers who suggested the possibility that the authors’ mixed-feature formulation might be equivalent to a bounded continuous-feature model, which directly inspired Section 5.2 in this manuscript. The authors are also indebted to Oscar Dowson for his invaluable guidance on efficient implementations of our Julia source codes.