SPOT: Scalable Policy Optimization with Trees for Markov Decision Processes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xiong, Xuyuan, Chumpitaz-Flores, Pedro, Hua, Kaixun, Hua, Cheng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912663842324480
author Xiong, Xuyuan
Chumpitaz-Flores, Pedro
Hua, Kaixun
Hua, Cheng
author_facet Xiong, Xuyuan
Chumpitaz-Flores, Pedro
Hua, Kaixun
Hua, Cheng
contents Interpretable reinforcement learning policies are essential for high-stakes decision-making, yet optimizing decision tree policies in Markov Decision Processes (MDPs) remains challenging. We propose SPOT, a novel method for computing decision tree policies, which formulates the optimization problem as a mixed-integer linear program (MILP). To enhance efficiency, we employ a reduced-space branch-and-bound approach that decouples the MDP dynamics from tree-structure constraints, enabling efficient parallel search. This significantly improves runtime and scalability compared to previous methods. Our approach ensures that each iteration yields the optimal decision tree. Experimental results on standard benchmarks demonstrate that SPOT achieves substantial speedup and scales to larger MDPs with a significantly higher number of states. The resulting decision tree policies are interpretable and compact, maintaining transparency without compromising performance. These results demonstrate that our approach simultaneously achieves interpretability and scalability, delivering high-quality policies an order of magnitude faster than existing approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2510_19241
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SPOT: Scalable Policy Optimization with Trees for Markov Decision Processes
Xiong, Xuyuan
Chumpitaz-Flores, Pedro
Hua, Kaixun
Hua, Cheng
Machine Learning
Artificial Intelligence
Interpretable reinforcement learning policies are essential for high-stakes decision-making, yet optimizing decision tree policies in Markov Decision Processes (MDPs) remains challenging. We propose SPOT, a novel method for computing decision tree policies, which formulates the optimization problem as a mixed-integer linear program (MILP). To enhance efficiency, we employ a reduced-space branch-and-bound approach that decouples the MDP dynamics from tree-structure constraints, enabling efficient parallel search. This significantly improves runtime and scalability compared to previous methods. Our approach ensures that each iteration yields the optimal decision tree. Experimental results on standard benchmarks demonstrate that SPOT achieves substantial speedup and scales to larger MDPs with a significantly higher number of states. The resulting decision tree policies are interpretable and compact, maintaining transparency without compromising performance. These results demonstrate that our approach simultaneously achieves interpretability and scalability, delivering high-quality policies an order of magnitude faster than existing approaches.
title SPOT: Scalable Policy Optimization with Trees for Markov Decision Processes
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.19241