Bellman goes Relational. Kersting, K.; Otterlo, M. V.; and Raedt, L. D. Proceedings of the Twenty-First International Conference on Machine Learning (ICML-2004), 2004.
abstract   bibtex   
Motivated by the interest in relational reinforcement learning, we introduce a novel relational Bellman update operator called ReBel. It employs a constraint logic programming language to compactly represent Markov decision processes over relational domains. Using ReBel, a novel value iteration algorithm is developed in which abstraction (over states and actions) plays a major role. This framework provides new insights into relational reinforcement learning. Convergence results as well as experiments are presented.
@article{Kersting:2004,
  abstract = {Motivated by the interest in relational reinforcement
	
	learning, we introduce a novel
	
	relational Bellman update operator called
	
	ReBel. It employs a constraint logic programming
	
	language to compactly represent
	
	Markov decision processes over relational domains.
	
	Using ReBel, a novel value iteration
	
	algorithm is developed in which abstraction
	
	(over states and actions) plays a major role.
	
	This framework provides new insights into relational
	
	reinforcement learning. Convergence
	
	results as well as experiments are presented.},
  author = {Kersting, Kristian and Otterlo, Martijn Van and Raedt, Luc De},
  interhash = {640ed3c7c9747d4da813c1bff4eaf47e},
  intrahash = {6811a3df8d0e864d523b5d6263262dcb},
  journal = {Proceedings of the Twenty-First International Conference on Machine
	Learning ({ICML}-2004)},
  pages = {465-472},
  title = {Bellman goes Relational},
  year = 2004
}
Downloads: 0