Discounted Markov decision processes via time aggregation.
Edilson F. ArrudaMarcelo D. FragosoPublished in: ECC (2016)
Keyphrases
- markov decision processes
- optimal policy
- finite state
- infinite horizon
- reinforcement learning
- state space
- finite horizon
- transition matrices
- policy iteration
- planning under uncertainty
- decision theoretic planning
- dynamic programming
- average reward
- markov decision process
- reachability analysis
- action space
- model based reinforcement learning
- partially observable
- factored mdps
- state and action spaces
- action sets
- total reward
- discount factor
- reinforcement learning algorithms
- reward function
- linear programming
- average cost