Distributional Random Forests: Heterogeneity Adjustment and Multivariate Distributional Regression
This method addresses the need for flexible distributional regression in machine learning, offering a novel approach for multivariate data analysis, though it appears incremental by extending random forest concepts.
The paper tackles the problem of estimating conditional distributions for multivariate responses using random forests, proposing a new splitting criterion based on the MMD metric to detect heterogeneity, and demonstrates versatility across various examples with available code.
Random Forest (Breiman, 2001) is a successful and widely used regression and classification algorithm. Part of its appeal and reason for its versatility is its (implicit) construction of a kernel-type weighting function on training data, which can also be used for targets other than the original mean estimation. We propose a novel forest construction for multivariate responses based on their joint conditional distribution, independent of the estimation target and the data model. It uses a new splitting criterion based on the MMD distributional metric, which is suitable for detecting heterogeneity in multivariate distributions. The induced weights define an estimate of the full conditional distribution, which in turn can be used for arbitrary and potentially complicated targets of interest. The method is very versatile and convenient to use, as we illustrate on a wide range of examples. The code is available as Python and R packages drf.