8289889263db4a40463e3f358bb7c7a1-AuthorFeedback.pdf

First of all, we thank the reviewers for their helpful comments and remarks. Moore-Penrose pseudo-inverse which explains why the complexity is only quadratic in K). The notation φ in Eq. 3 means that φ is a multivariate function of the p variables Reviewer 1. Ooops, you are right: The expectation in the expression of a Reviewer 2. An important difference wrt to the work by Gérard Biau, Luc Devroye and Gabor Lugosi ([1]) is that we Adaptative Neural Trees ([2]) and Deep Neural Decision Forests ([3]) are both built from decision trees. Reviewer 3. It is true that a standard regression tree with enough leaves can also approximate a smooth link function. We'll modify lines 228-230 as Reviewer 4. One can obtain standard regression trees from Eqs 2 and 3 by setting φ to (2π) In that case, the distribution of x over regions is concentrated on one region.