On the Optimality of Sparse Model-Based Planning for Markov Decision Processes.
Alekh AgarwalSham M. KakadeLin F. YangPublished in: CoRR (2019)
Keyphrases
- markov decision processes
- average cost
- macro actions
- average reward
- decision theoretic planning
- planning under uncertainty
- partially observable
- finite state
- interval estimation
- optimal policy
- state space
- reinforcement learning
- reachability analysis
- policy iteration
- transition matrices
- infinite horizon
- stationary policies
- heuristic search
- partially observable markov decision processes
- decision processes
- dynamic programming
- factored mdps
- finite horizon
- markov decision problems
- probabilistic planning
- planning problems
- model free
- action space
- initial state
- model based reinforcement learning
- optimal solution
- reinforcement learning algorithms
- ai planning
- state abstraction
- action sets
- optimality criterion
- risk sensitive
- sufficient conditions
- dynamical systems
- state and action spaces
- least squares
- reward function
- multistage
- planning graph
- markov decision process
- semi markov decision processes