Proto-value Functions: A Laplacian Framework for Learning Representation and Control in Markov Decision Processes.
Sridhar MahadevanMauro MaggioniPublished in: J. Mach. Learn. Res. (2007)
Keyphrases
- markov decision processes
- decision theoretic planning
- reinforcement learning
- learning algorithm
- model based reinforcement learning
- reinforcement learning algorithms
- supervised learning
- partially observable
- stochastic games
- control system
- learning tasks
- state space
- multistage
- infinite horizon
- policy iteration
- decentralized control
- dynamic programming
- factored mdps