Between MDPs and Semi-MDPs: A Framework for Temporal Abstraction in Reinforcement Learning


Learning, planning, and representing knowledge at multiple levels of temporal abstraction are key, longstanding challenges for AI. In this paper we consider how these challenges can be addressed within the mathematical framework of reinforcement learning and Markov decision processes (MDPs). We extend the usual notion of action in this framework to include… (More)
DOI: 10.1016/S0004-3702(99)00052-1



Citations per Year

1,565 Citations

Semantic Scholar estimates that this publication has 1,565 citations based on the available data.

See our FAQ for additional information.