GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Marton, Sascha, Lüdtke, Stefan, Bartelt, Christian, Stuckenschmidt, Heiner
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910570332028928
author Marton, Sascha
Lüdtke, Stefan
Bartelt, Christian
Stuckenschmidt, Heiner
author_facet Marton, Sascha
Lüdtke, Stefan
Bartelt, Christian
Stuckenschmidt, Heiner
contents Decision Trees (DTs) are commonly used for many machine learning tasks due to their high degree of interpretability. However, learning a DT from data is a difficult optimization problem, as it is non-convex and non-differentiable. Therefore, common approaches learn DTs using a greedy growth algorithm that minimizes the impurity locally at each internal node. Unfortunately, this greedy procedure can lead to inaccurate trees. In this paper, we present a novel approach for learning hard, axis-aligned DTs with gradient descent. The proposed method uses backpropagation with a straight-through operator on a dense DT representation, to jointly optimize all tree parameters. Our approach outperforms existing methods on binary classification benchmarks and achieves competitive results for multi-class tasks. The method is available under: https://github.com/s-marton/GradTree
format Preprint
id arxiv_https___arxiv_org_abs_2305_03515
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent
Marton, Sascha
Lüdtke, Stefan
Bartelt, Christian
Stuckenschmidt, Heiner
Machine Learning
Artificial Intelligence
Decision Trees (DTs) are commonly used for many machine learning tasks due to their high degree of interpretability. However, learning a DT from data is a difficult optimization problem, as it is non-convex and non-differentiable. Therefore, common approaches learn DTs using a greedy growth algorithm that minimizes the impurity locally at each internal node. Unfortunately, this greedy procedure can lead to inaccurate trees. In this paper, we present a novel approach for learning hard, axis-aligned DTs with gradient descent. The proposed method uses backpropagation with a straight-through operator on a dense DT representation, to jointly optimize all tree parameters. Our approach outperforms existing methods on binary classification benchmarks and achieves competitive results for multi-class tasks. The method is available under: https://github.com/s-marton/GradTree
title GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2305.03515