Parallel Monte Carlo Tree Search from Multi-core to Many-core Processors

Mirsoleimani, S. Ali; Plaat, Aske; Herik, Jaap van den; Vermaseren, J.A.M.

doi:10.1109/trustcom.2015.615

Cited by 9 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The plain UCT algorithm and Ensemble UCT are implemented in C++. In order to make our experiments as realistic as possible, we use a custom developed game playing program for the game of Hex (Mirsoleimani et al, 2014;Mirsoleimani et al, 2015). This program is highly optimized, and reaches a speed of more than 40,000 playouts per second per core on a 2,4 GHz Intel Xeon processor.…”

Section: Methodsmentioning

confidence: 99%

Ensemble UCT Needs High Exploitation

Mirsoleimani

Plaat

Herik

2016

Proceedings of the 8th International Conference on Agents and Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

Abstract:Recent results have shown that the MCTS algorithm (a new, adaptive, randomized optimization algorithm) is effective in a remarkably diverse set of applications in Artificial Intelligence, Operations Research, and High Energy Physics. MCTS can find good solutions without domain dependent heuristics, using the UCT formula to balance exploitation and exploration. It has been suggested that the optimum in the exploitationexploration balance differs for different search tree sizes: small search trees needs more exploitation; large search trees need more exploration. Small search trees occur in variations of MCTS, such as parallel and ensemble approaches. This paper investigates the possibility of improving the performance of Ensemble UCT by increasing the level of exploitation. As the search trees becomes smaller we achieve an improved performance. The results are important for improving the performance of large scale parallelism of MCTS.

show abstract

Section: Methodsmentioning

confidence: 99%

Ensemble UCT Needs High Exploitation

Mirsoleimani

Plaat

Herik

2016

Proceedings of the 8th International Conference on Agents and Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

show abstract

“…Existing parallel methods use iteration-level (IL) parallelism. They assign each iteration to a separate processing element (thread) for execution on separate processors [5], [8], [9]. All three publications are facing a bottleneck in their implementation since they can not partition the iteration into constituent parts (operations) for parallel execution.…”

Section: Introductionmentioning

confidence: 99%

Structured Parallel Programming for Monte Carlo Tree Search

Mirsoleimani¹,

Plaat²,

Herik³

et al. 2017

Preprint

Self Cite

View full text Add to dashboard Cite

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. Moreover, we propose a new lock-free tree data structure for parallel MCTS which removes synchronization overhead. The Pipeline Pattern for Parallel MCTS algorithm (called 3PMCTS), scales very well to higher numbers of cores when compared to the existing methods.

show abstract

“…A significant research effort has been put in the development of parallel algorithms for Monte Carlo Tree Search (MCTS) [1], [2], [3]. Parallel algorithms for MCTS come in multiple versions [4].…”

Section: Introductionmentioning

confidence: 99%

“…It is also shown that playing strength of MCTS will decrease to some degree with more parallel processors due to search overhead [7]. It is also shown that on many-core processors, perfect playout-speedup is really hard to achieve due to synchronization overhead [3].…”

Section: Introductionmentioning

confidence: 99%

A New Method for Parallel Monte Carlo Tree Search

Mirsoleimani,

Plaat,

Herik

et al. 2016

Preprint

Self Cite

View full text Add to dashboard Cite

In recent years there has been much interest in the Monte Carlo tree search algorithm, a new, adaptive, randomized optimization algorithm. In fields as diverse as Artificial Intelligence, Operations Research, and High Energy Physics, research has established that Monte Carlo tree search can find good solutions without domain dependent heuristics. However, practice shows that reaching high performance on large parallel machines is not so successful as expected. This paper proposes a new method for parallel Monte Carlo tree search based on the pipeline computation pattern.

show abstract

Parallel Monte Carlo Tree Search from Multi-core to Many-core Processors

Cited by 9 publications

References 13 publications

Ensemble UCT Needs High Exploitation

Ensemble UCT Needs High Exploitation

Structured Parallel Programming for Monte Carlo Tree Search

A New Method for Parallel Monte Carlo Tree Search

Contact Info

Product

Resources

About