.png){kind=icon}
.png){kind=icon}
Return
Soft regression trees: A model variant and a decomposition training algorithm
.png){kind=icon}
OA
AI
DOI：10.1016/j.ejor.2025.08.050.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A new soft regression trees variant with the conditional computation property is proposed. • A universal approximation result is provided for such soft regression trees. • A convergent training decomposition general scheme is presented. • A convergent version with initialization and reassignment heuristics is developed. • Our algorithm outperforms state-of-the-art deterministic/soft regression tree methods.
Journal
IF:
6
Papers:
2.2W
Citations:
6.4W