The efficiency of alpha-beta search algorithms heavily depends on the order in which the moves are examined. This paper focuses on using neural networks to estimate the likelihood of a move being the best in a certain position. The moves considered more likely to be the best are examined first. We selected Lines of Action as a testing ground. We investigated several schemes to encode the moves in a neural net- work. In the experiments, the best performance was obtained by using one output unit for each possible move of the game. The results indicate that our move-ordering approach can speed up the search with 20 to 50 percent compared with one of the best current alternatives, the history heuristic. © 2001 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Kocsis, L., Uiterwijk, J., & Van Den Herik, J. (2001). Move ordering using neural networks. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2070, 45–50. https://doi.org/10.1007/3-540-45517-5_6
Mendeley helps you to discover research relevant for your work.