Saved in:
Bibliographic Details
Main Authors: Fleischer, David, Stephens, David A., Yang, Archer Y.
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.11908
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915347153551360
author Fleischer, David
Stephens, David A.
Yang, Archer Y.
author_facet Fleischer, David
Stephens, David A.
Yang, Archer Y.
contents We propose a computationally efficient alternative to generalized random forests (GRFs) for estimating heterogeneous effects in large dimensions. While GRFs rely on a gradient-based splitting criterion, which in large dimensions is computationally expensive and unstable, our method introduces a fixed-point approximation that eliminates the need for Jacobian estimation. This gradient-free approach preserves GRF's theoretical guarantees of consistency and asymptotic normality while significantly improving computational efficiency. We demonstrate that our method achieves a speedup of multiple times over standard GRFs without compromising statistical accuracy. Experiments on both simulated and real-world data validate our approach. Our findings suggest that the proposed method is a scalable alternative for localized effect estimation in machine learning and causal inference applications
format Preprint
id arxiv_https___arxiv_org_abs_2306_11908
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Generalized Random Forests using Fixed-Point Trees
Fleischer, David
Stephens, David A.
Yang, Archer Y.
Machine Learning
Methodology
We propose a computationally efficient alternative to generalized random forests (GRFs) for estimating heterogeneous effects in large dimensions. While GRFs rely on a gradient-based splitting criterion, which in large dimensions is computationally expensive and unstable, our method introduces a fixed-point approximation that eliminates the need for Jacobian estimation. This gradient-free approach preserves GRF's theoretical guarantees of consistency and asymptotic normality while significantly improving computational efficiency. We demonstrate that our method achieves a speedup of multiple times over standard GRFs without compromising statistical accuracy. Experiments on both simulated and real-world data validate our approach. Our findings suggest that the proposed method is a scalable alternative for localized effect estimation in machine learning and causal inference applications
title Generalized Random Forests using Fixed-Point Trees
topic Machine Learning
Methodology
url https://arxiv.org/abs/2306.11908