Title :
Exemplar-based direct policy search with evolutionary optimization
Author_Institution :
Acad. Center for Comput. & Media Studies, Kyoto Univ., Japan
Abstract :
In this paper, an exemplar-based policy optimization framework for direct policy search is presented. In this exemplar-based approach, the policy to be optimized is composed of a set of exemplars and a case-based action selector. An implementation of this approach using a state-action-based policy representation and an evolutionary algorithm optimizer is shown to provide favorable search performance for two higher-dimensional problems.
Keywords :
evolutionary computation; learning by example; search problems; case-based action selector; direct policy search; evolutionary algorithm; evolutionary optimization; exemplar-based policy optimization; higher-dimensional problems; policy representation; search performance; state action; Artificial neural networks; Concrete; Evolutionary computation; Machine learning; Machine learning algorithms; Magnetic heads; PD control; Proportional control; State-space methods;
Conference_Titel :
Evolutionary Computation, 2005. The 2005 IEEE Congress on
Print_ISBN :
0-7803-9363-5
DOI :
10.1109/CEC.2005.1554988
