Exploration exploitation in Go: UCT for Monte-Carlo Go

By Sylvain Gelly, Yizao Wang. 2006

238 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Episodic Exploration for Deep Deterministic Policies: An Application to StarCraft Micromanagement Tasks (2016, methodology): “In RTS games like StarCraft, we do not have access to a simulator (and writing one would be a daunting task), so we cannot use (Monte Carlo) tree search [10] directly, even less so in the setting of full games [23].”
On the Analysis of Complex Backup Strategies in Monte Carlo Tree Search (2016, unknown): “Various modifications have been proposed to improve the exploration strategy under UCT (Gelly and Wang, 2006; Wang and Gelly, 2007) and its value estimation (Gelly and Silver, 2007; 2011; Helmbold and Parker- Wood, 2009; Feldman and Domshlak, 2014b).”
Solving two-armed Bernoulli bandit problems using a Bayesian learning automaton (2010, methodology): “For instance, UCB-tuned (Auer et al., 2002) is used for move exploration in MoGo, a top-level computer-go program on 9 £ 9 go boards (Gelly and Wang, 2006).”
Monte Carlo Tree Search: a review of recent modifications and applications (2021, background): “ 3 from the work by Gelly and Wang (2006).”
Variance Reduction in Monte-Carlo Tree Search (2011, unknown): “Noteworthy enhancements include the addition of domain knowledge [12, 13], parallelization [7], Rapid Action Value Estimation (RAVE) [11], automated parameter tuning [8] and rollout policy optimization [21].”
Parallel Monte-Carlo Tree Search (2008, background): “Nevertheless, in 2006 Monte-Carlo Tree Search (MCTS) [7, 4, 12, 8, 10, 6] started a revolution in the ﬁeld of Computer Go.”
TORCS, The Open Racing Car Simulator (2005, background): “The overall racing problem itself can be formalised as maximising the probability of winning a race, minimising expected race rank or maximising the expected number of points obtained for the team in a championship setting.”
Progressive Strategies for Monte-Carlo Tree Search (2008, background): “Two-person zero-sum games with perfect information have been addressed by many AI researchers with great success for fifty years [van den Herik et al. (2002)].”
Empirical evaluation of ad hoc teamwork in the pursuit domain (2011, methodology): “UCT has been shown to be effective in games with a high branching factor, such as Go [7], so it should be able to handle the branching factor caused by the number of agents.”
A Survey of Monte Carlo Tree Search Methods (2012, methodology): “TDL can learn heuristic value functions to inform the tree policy or the simulation (playout) policy.”