Approximately-Optimal Queries for Planning in Reward-Uncertain Markov Decision Processes.
Shun ZhangEdmund H. DurfeeSatinder P. SinghPublished in: ICAPS (2017)
Keyphrases
- markov decision processes
- approximately optimal
- average reward
- reinforcement learning
- reward function
- macro actions
- planning under uncertainty
- decision theoretic planning
- partially observable
- optimal policy
- discounted reward
- state space
- expected reward
- policy iteration
- transition matrices
- total reward
- finite state
- infinite horizon
- reinforcement learning algorithms
- decision making
- partially observable markov decision processes
- dynamic programming
- stationary policies
- long run
- markov decision problems
- probabilistic planning
- average cost
- action space
- finite horizon
- planning problems
- approximation ratio
- state and action spaces
- markov decision process
- mechanism design
- special case
- partially observed
- utility function
- heuristic search