Pipeline Pattern for Parallel MCTS

S. Ali Mirsoleimani, Jaap van den Herik, Aske Plaat, Jos Vermaseren

2018

Abstract

In this paper, we present a new algorithm for parallel Monte Carlo tree search (MCTS). It is based on the pipeline pattern and allows flexible management of the control flow of the operations in parallel MCTS. The pipeline pattern provides for the first structured parallel programming approach to MCTS. The Pipeline Pattern for Parallel MCTS algorithm (called 3PMCTS) scales very well to a higher number of cores when compared to the existing methods. The observed speedup is 21 on a 24-core machine.

Paper Citation

in Harvard Style

Mirsoleimani S., van den Herik J., Plaat A. and Vermaseren J. (2018). Pipeline Pattern for Parallel MCTS.In Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-275-2, pages 614-621. DOI: 10.5220/0006656306140621

in Bibtex Style

@conference{icaart18,
author={S. Ali Mirsoleimani and Jaap van den Herik and Aske Plaat and Jos Vermaseren},
title={Pipeline Pattern for Parallel MCTS},
booktitle={Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2018},
pages={614-621},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006656306140621},
isbn={978-989-758-275-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Pipeline Pattern for Parallel MCTS
SN - 978-989-758-275-2
AU - Mirsoleimani S.
AU - van den Herik J.
AU - Plaat A.
AU - Vermaseren J.
PY - 2018
SP - 614
EP - 621
DO - 10.5220/0006656306140621