Surrogate Assisted Monte Carlo Tree Search in Combinatorial Optimization
Abstract
Industries frequently adjust their facilities network by opening new branches in promising areas and closing branches in areas where they expect low profits. In this paper, we examine a particular class of facility location problems. Our objective is to minimize the loss of sales resulting from the removal of several retail stores. However, estimating sales accurately is expensive and time-consuming. To overcome this challenge, we leverage Monte Carlo Tree Search (MCTS) assisted by a surrogate model that computes evaluations faster. Results suggest that MCTS supported by a fast surrogate function can generate solutions faster while maintaining a consistent solution compared to MCTS that does not benefit from the surrogate function.
Keywords
Cite
@article{arxiv.2403.09925,
title = {Surrogate Assisted Monte Carlo Tree Search in Combinatorial Optimization},
author = {Saeid Amiri and Parisa Zehtabi and Danial Dervovic and Michael Cashmore},
journal= {arXiv preprint arXiv:2403.09925},
year = {2024}
}
Comments
Accepted to the ICAPS Planning and Scheduling for Financial Services (FINPLAN) 2023 workshop