Measuring the Distance Between Finite Markov Decision Processes.
Jinhua SongYang GaoHao WangBo AnPublished in: AAMAS (2016)
Keyphrases
- markov decision processes
- state and action spaces
- finite state
- reinforcement learning
- state space
- optimal policy
- transition matrices
- policy iteration
- dynamic programming
- average reward
- reachability analysis
- finite horizon
- decision theoretic planning
- reinforcement learning algorithms
- planning under uncertainty
- model based reinforcement learning
- risk sensitive
- action space
- stationary policies
- markov decision process
- factored mdps
- partially observable
- infinite horizon
- finite number
- interval estimation
- markov decision problems
- real time dynamic programming
- machine learning
- decision diagrams
- average cost
- action sets
- semi markov decision processes
- linear programming