Abstract
Monte-Carlo tree search (MCTS) is a new technique that has produced a huge leap forward in the strength of Go-playing programs. An interesting aspect of MCTS that has been rarely studied in the past is the problem of time management. This paper presents the effect on playing strength of a variety of time-management heuristics for 19 × 19 Go. Results indicate that clever time management can have a very significant effect on playing strength. Experiments demonstrate that the most basic algorithm for sudden-death time controls (dividing the remaining time by a constant) produces a winning rate of 43.2±2.2% against GNU Go 3.8 Level 2, whereas our most efficient time-allocation strategy can reach a winning rate of 60±2.2% without pondering and 67.4±2.1% with pondering.
Original language | English |
---|---|
Title of host publication | Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010 |
Pages | 462-466 |
Number of pages | 5 |
DOIs | |
Publication status | Published - 2010 Dec 1 |
Event | 2010 15th Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010 - Hsinchu, Taiwan Duration: 2010 Nov 18 → 2010 Nov 20 |
Publication series
Name | Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010 |
---|
Other
Other | 2010 15th Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010 |
---|---|
Country | Taiwan |
City | Hsinchu |
Period | 10/11/18 → 10/11/20 |
Fingerprint
Keywords
- Game of Go
- Monte-Carlo tree search
- Time management
ASJC Scopus subject areas
- Artificial Intelligence
- Computational Theory and Mathematics
Cite this
Time management for Monte-Carlo tree search applied to the game of Go. / Huang, Shih Chieh; Coulom, Remi; Lin, Shun-Shii.
Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010. 2010. p. 462-466 5695493 (Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Time management for Monte-Carlo tree search applied to the game of Go
AU - Huang, Shih Chieh
AU - Coulom, Remi
AU - Lin, Shun-Shii
PY - 2010/12/1
Y1 - 2010/12/1
N2 - Monte-Carlo tree search (MCTS) is a new technique that has produced a huge leap forward in the strength of Go-playing programs. An interesting aspect of MCTS that has been rarely studied in the past is the problem of time management. This paper presents the effect on playing strength of a variety of time-management heuristics for 19 × 19 Go. Results indicate that clever time management can have a very significant effect on playing strength. Experiments demonstrate that the most basic algorithm for sudden-death time controls (dividing the remaining time by a constant) produces a winning rate of 43.2±2.2% against GNU Go 3.8 Level 2, whereas our most efficient time-allocation strategy can reach a winning rate of 60±2.2% without pondering and 67.4±2.1% with pondering.
AB - Monte-Carlo tree search (MCTS) is a new technique that has produced a huge leap forward in the strength of Go-playing programs. An interesting aspect of MCTS that has been rarely studied in the past is the problem of time management. This paper presents the effect on playing strength of a variety of time-management heuristics for 19 × 19 Go. Results indicate that clever time management can have a very significant effect on playing strength. Experiments demonstrate that the most basic algorithm for sudden-death time controls (dividing the remaining time by a constant) produces a winning rate of 43.2±2.2% against GNU Go 3.8 Level 2, whereas our most efficient time-allocation strategy can reach a winning rate of 60±2.2% without pondering and 67.4±2.1% with pondering.
KW - Game of Go
KW - Monte-Carlo tree search
KW - Time management
UR - http://www.scopus.com/inward/record.url?scp=79951753159&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79951753159&partnerID=8YFLogxK
U2 - 10.1109/TAAI.2010.78
DO - 10.1109/TAAI.2010.78
M3 - Conference contribution
AN - SCOPUS:79951753159
SN - 9780769542539
T3 - Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010
SP - 462
EP - 466
BT - Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010
ER -