Symbolic Dynamic Programming for Discrete and Continuous State MDPs.
Scott SannerKarina Valdivia DelgadoLeliane Nunes de BarrosPublished in: UAI (2011)
Keyphrases
- continuous state
- dec pomdps
- dynamic programming
- reinforcement learning
- markov decision processes
- continuous state and action spaces
- policy search
- continuous state spaces
- action space
- state space
- finite state
- continuous action
- partially observable markov decision processes
- robot navigation
- state dependent
- optimal policy
- control policies
- planning problems
- markov decision problems
- state action
- markov chain
- action selection
- state transition