Optimistic planning for Markov decision processes.
Lucian BusoniuRémi MunosPublished in: AISTATS (2012)
Keyphrases
- markov decision processes
- macro actions
- decision theoretic planning
- planning under uncertainty
- partially observable
- state space
- transition matrices
- finite state
- optimal policy
- dynamic programming
- probabilistic planning
- reinforcement learning
- factored mdps
- average reward
- policy iteration
- model based reinforcement learning
- reinforcement learning algorithms
- partially observable markov decision processes
- planning problems
- markov decision problems
- reachability analysis
- finite horizon
- infinite horizon
- decision processes
- ai planning
- risk sensitive
- heuristic search
- domain independent
- average cost
- blocks world
- markov decision process
- classical planning
- discounted reward
- semi markov decision processes
- stochastic shortest path
- action sets
- stationary policies
- machine learning
- markov chain
- least squares
- learning algorithm