Título: Improved automatic discovery of subgoals for options in hierarchical
Autores: Kretchmar, R. Matthew
Feil, Todd
Bansal, Rohit
Fecha: 2004-05-17
2003
Publicador: Unversidad Nacional de La Plata
Fuente:


Tipo: Articulo
Articulo
Tema: reinforcement learning; options; subgoal discovery
Ciencias Informáticas
Reinforcement Learning
Inteligencia artificial
Descripción: Options have been shown to be a key step in extending reinforcement learning beyond low-level reactionary systems to higher-level, planning systems. Most of the options research involves hand-crafted options; there has been only very limited work in the automated discovery of options. We extend early work in automated option discovery with a flexible and robust method.
Idioma: Inglés