TY - GEN
T1 - Fast seed-learning algorithms for games
AU - LIU, Jialin
AU - TEYTAUD, Olivier
AU - CAZENAVE, Tristan
N1 - Publisher Copyright:
© Springer International Publishing AG 2016.
PY - 2016
Y1 - 2016
N2 - Recently, a methodology has been proposed for boosting the computational intelligence of randomized game-playing programs. We propose faster variants of these algorithms, namely rectangular algorithms (fully parallel) and bandit algorithms (faster in a sequential setup). We check the performance on several board games and card games. In addition, in the case of Go, we check the methodology when the opponent is completely distinct to the one used in the training.
AB - Recently, a methodology has been proposed for boosting the computational intelligence of randomized game-playing programs. We propose faster variants of these algorithms, namely rectangular algorithms (fully parallel) and bandit algorithms (faster in a sequential setup). We check the performance on several board games and card games. In addition, in the case of Go, we check the methodology when the opponent is completely distinct to the one used in the training.
UR - http://www.scopus.com/inward/record.url?scp=85007314942&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-50935-8_6
DO - 10.1007/978-3-319-50935-8_6
M3 - Conference paper (refereed)
AN - SCOPUS:85007314942
SN - 9783319509341
T3 - Lecture Notes in Computer Science
SP - 58
EP - 70
BT - Computers and Games : 9th International Conference, CG 2016, Revised Selected Papers
A2 - PLAAT, Aske
A2 - KOSTERS, Walter
A2 - VAN DEN HERIK, Jaap
PB - Springer-Verlag Italia Srl
T2 - 9th International Conference on Computer and Games, CG 2016
Y2 - 29 June 2016 through 1 July 2016
ER -