Keyphrases
- optimal policy
- markov decision processes
- stationary policies
- markov decision process
- markov decision problems
- reward function
- reinforcement learning
- policy search
- average cost
- state space
- stochastic domains
- policy iteration
- finite horizon
- factored mdps
- finite state
- dynamic programming
- partially observable markov decision processes
- average reward
- decision processes
- factored markov decision processes
- decision diagrams
- black box
- decision theoretic planning
- partially observable
- semi markov decision processes
- linear programming
- sufficient conditions
- multistage
- decision problems
- infinite horizon
- decentralized control
- initial state
- utility function
- hierarchical reinforcement learning
- control policies
- expected reward
- discounted reward
- real time dynamic programming
- transition probabilities