Learned Look-Ahead Splitting Rule for CART
Abstract
Classification and regression trees are typically constructed using a greedy splitting rule that maximizes the immediate reduction in prediction error at each node. Although this strategy is computationally efficient, it can miss splits that yield small short-term gains but create substantial downstream improvements after further partitioning. We propose a look-ahead tree-building method that evaluates each candidate split by the prediction error reduction achieved after growing a conventional CART subtree below that split. Because the full look-ahead procedure can be computationally expensive, we also describe a smart look-ahead algorithm that learns downstream split values using node-level features. The proposed framework preserves the interpretability of recursive partitioning while improving split selection in hierarchical or interaction-driven settings. We conduct a simulation study comparing conventional, full look-ahead, and smart look-ahead methods under several settings and apply the proposed methods to analyze two real data examples demonstrating the merit of the new methods.
Related
- LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators
- Mean squared error reduction of (plug-in) look-ahead estimators for Markov chains
- The Hidden Look-Ahead: Point-in-Time Leakage in Electricity Market Data
- Risk-Aware Look-Ahead Planning for Independent Multi-Goal UAV Paths: A Bounded Two-Dimensional Geometric Study
- Weighting-Based Strategies for Classification: Analysis and Local Peak Dynamics with CART Trees
BibTeX
@misc{gao2026learned,
title = {{Learned Look-Ahead Splitting Rule for CART}},
author = {Andrew Gao and Tianlin Liu and Ruichen Han and Lu Tian},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.16440},
eprint = {2609.16440},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.16440},
}
RIS
TY - UNPB
TI - Learned Look-Ahead Splitting Rule for CART
AU - Andrew Gao
AU - Tianlin Liu
AU - Ruichen Han
AU - Lu Tian
PY - 2026
DA - 2026/09/14
PB - arXiv
DO - 10.48550/arxiv.2609.16440
UR - https://arxiv.org/abs/2609.16440
AB - Classification and regression trees are typically constructed using a greedy splitting rule that maximizes the immediate reduction in prediction error at each node. Although this strategy is computationally efficient, it can miss splits that yield small short-term gains but create substantial downstream improvements after further partitioning. We propose a look-ahead tree-building method that evaluates each candidate split by the prediction error reduction achieved after growing a conventional CART subtree below that split. Because the full look-ahead procedure can be computationally expensive, we also describe a smart look-ahead algorithm that learns downstream split values using node-level features. The proposed framework preserves the interpretability of recursive partitioning while improving split selection in hierarchical or interaction-driven settings. We conduct a simulation study comparing conventional, full look-ahead, and smart look-ahead methods under several settings and apply the proposed methods to analyze two real data examples demonstrating the merit of the new methods.
ER -