Differentiable Decision Tree via "ReLU+Argmin" Reformulation

NeurIPSSpotlight2025

Authors
Qiangqiang Mao, Jiayang Ren, Yixiu Wang, Chenxuanyin Zou, Jingjing Zheng, Yankai Cao
Venue
NeurIPS 2025
Track
Spotlight

TL;DR

Decision tree, despite its unmatched interpretability and lightweight structure, faces two key issues that limit its broader applicability: non-differentiability and low testing accuracy…

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

interpretability

← All NeurIPS 2025 Spotlight papers · Browse the whole archive