Achieving master level play in 9×9 computer go

By Sylvain Gelly, David Silver. 2008

129 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

X-Armed Bandits (2010, unknown): “al guarantees, we expect our algorithm to work well in practice since the algorithm is very close to the recent, empirically very successful tree-search methods from the games and planning literature [16; 17; 27; 11; 15]. Outline. The outline of the paper is as follows: 1. In Section 2 we formalize ”
Learning non-myopically from human-generated reward (2013, methodology): “UCT-based search has been successful in tasks with especially large state spaces [8], originally in games like Go [2] but also in more general reinforcement learning tasks [3].”
Guiding Combinatorial Optimization with UCT (2012, background): “It has received much attention during the past few years due to its success in game playing agents, especially for Go [4, 5] and Kriegspiel [2], as well as for general game playing [3].”
Integrated task and motion planning in belief space (2013, background): “The process of constructing and evaluating a tree of this kind can be prohibitively expensive; but there have been recent advances in effective sample-based approaches (Gelly and Silver, 2008).”
Multi-armed bandits with episode context (2011, methodology): “Computer Go has been very challenging [5, 9], but major advances have been obtained using Monte Carlo techniques that evaluate positions using random playouts [7, 19, 20 ].”
Policy evaluation with temporal differences: a survey and comparison (2015, background): “Policy-iteration-based reinforcement learning has yielded impressive applications in robot soccer (Riedmiller and Gabel, 2007), elevator control (Crites and Barto, 1998) and game-playing such as Checkers (Samuel, 1959), Backgammon (Tesauro, 1994) and Go (Gelly and Silver, 2008).”