45 search results - page 6 / 9 » Cross-Entropy for Monte-Carlo Tree Search |
JAIR
2011
13 years 4 months ago
2011
This paper describes a computationally feasible approximation to the AIXI agent, a universal reinforcement learning agent for arbitrary environments. AIXI is scaled down in two ke...
NLP
2000
14 years 1 months ago
2000
Abstract. The purpose of this paper is (1) to provide a theoretical justification for the use of Monte-Carlo sampling for approximate resolution of NP-hard maximization problems in...
KI
2011
Springer
13 years 4 months ago
2011
Springer
Abstract We present the General Game Playing system Centurio. Centurio is a Java-based player featuring different strategies based on Monte Carlo Tree Search extended by technique...
ACG
2003
Springer
14 years 3 months ago
2003
Springer
We describe two Go programs, ¢¡¤£¦¥ and ¢¡¤§¨£ , developed by a Monte-Carlo approach that is simpler than Bruegmann’s (1993) approach. Our method is based on Abra...
ACG
2009
Springer
14 years 4 months ago
2009
Springer
Recently, Monte-Carlo Tree Search (MCTS) has advanced the field of computer Go substantially. In the game of Lines of Action (LOA), which has been dominated in the past by αβ, M...