espace de recherche de la distribution gaussienne est plus expressif et le processus de recherche peut être plus efficace. Nous examinons actuellement l'intérêt d'utiliser CMA ,
An Evolutionary Approach to Tetris, The Sixth Metaheuristics International Conference (MIC2005), 2005. ,
How to Lose at Tetris, Mathematical Gazette, pp.194-200, 1997. ,
DOI : 10.2307/3619195
A Tutorial on the Cross-Entropy Method, Annals of Operations Research, vol.16, issue.3, pp.134-134, 2004. ,
DOI : 10.1007/s10479-005-5724-z
Tetris is Hard, Even to Approximate, Proc. 9th International Computing and Combinatorics Conference, pp.351-363, 2003. ,
DOI : 10.1007/3-540-45071-8_36
Tetris: A Study of Randomized Constraint Sampling, 2006. ,
DOI : 10.1007/1-84628-095-8_6
Feature Discovery in Reinforcement Learning Using Genetic Programming, 2007. ,
DOI : 10.1007/978-3-540-78671-9_19
URL : https://hal.archives-ouvertes.fr/hal-00826056
Completely Derandomized Self-Adaptation in Evolution Strategies, Evolutionary Computation, vol.9, issue.2, pp.159-195, 2001. ,
DOI : 10.1016/0004-3702(95)00124-7
Least-Squares Methods in Reinforcement Learning for Control, SETN '02 : Proceedings of the Second Hellenic Conference on AI, pp.249-260, 2002. ,
DOI : 10.1007/3-540-46014-4_23
« On the numeric stability of gaussian processes regression for relational reinforcement learning, Workshop on Relational Reinforcement Learning, pp.10-14, 2004. ,
An introduction, 1998. ,
Learning Tetris Using the Noisy Cross-Entropy Method, Neural Computation, vol.18, issue.12, pp.2936-2941, 2006. ,
DOI : 10.1007/s10479-005-5732-z
« Feature-Based Methods for Large Scale Dynamic Programming, Machine Learning, pp.59-94, 1996. ,