Ensemble UCT Needs High Exploitation

作者: S. Ali Mirsoleimani , Aske Plaat , Jaap van den Herik

DOI:

关键词:

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

参考文章(12)
H. Jaap van den Herik, Jos Vermaseren, S. Ali Mirsoleimani, Aske Plaat, Performance analysis of a 240 thread tournament level MCTS Go program on the Intel Xeon Phi arXiv: Performance. ,(2014)
Alan Fern, Paul Lewis, Ensemble Monte-Carlo planning: an empirical study international conference on automated planning and scheduling. pp. 58- 65 ,(2011)
J.W. Romein, Multigame - An Environment for Distributed Game- Tree Search Amsterdam: Vrije Universiteit. ,(2001)
Guillaume M. J. -B. Chaslot, Mark H. M. Winands, H. Jaap van den Herik, Parallel Monte-Carlo Tree Search Computers and Games. pp. 60- 71 ,(2008) , 10.1007/978-3-540-87608-3_6
Rémi Coulom, Efficient selectivity and backup operators in Monte-Carlo tree search annual conference on computers. pp. 72- 83 ,(2006) , 10.1007/978-3-540-75538-8_7
GUILLAUME M. J-B. CHASLOT, MARK H. M. WINANDS, H. JAAP VAN DEN HERIK, JOS W. H. M. UITERWIJK, BRUNO BOUZY, Progressive Strategies for Monte-Carlo Tree Search New Mathematics and Natural Computation. ,vol. 4, pp. 343- 357 ,(2008) , 10.1142/S1793005708001094
Zvi Galil, Giuseppe F. Italiano, Data structures and algorithms for disjoint set union problems ACM Computing Surveys. ,vol. 23, pp. 319- 344 ,(1991) , 10.1145/116873.116878
Broderick Arneson, Ryan B. Hayward, Philip Henderson, Monte Carlo Tree Search in Hex IEEE Transactions on Computational Intelligence and AI in Games. ,vol. 2, pp. 251- 258 ,(2010) , 10.1109/TCIAIG.2010.2067212
Cameron B. Browne, Edward Powley, Daniel Whitehouse, Simon M. Lucas, Peter I. Cowling, Philipp Rohlfshagen, Stephen Tavener, Diego Perez, Spyridon Samothrakis, Simon Colton, A Survey of Monte Carlo Tree Search Methods IEEE Transactions on Computational Intelligence and AI in Games. ,vol. 4, pp. 1- 43 ,(2012) , 10.1109/TCIAIG.2012.2186810
Yusuke Soejima, Akihiro Kishimoto, Osamu Watanabe, Evaluating Root Parallelization in Go IEEE Transactions on Computational Intelligence and AI in Games. ,vol. 2, pp. 278- 287 ,(2010) , 10.1109/TCIAIG.2010.2096427