Keyphrases
- markov decision processes
- optimal policy
- discounted reward
- policy iteration
- markov decision process
- average reward
- finite horizon
- infinite horizon
- average cost
- state and action spaces
- partially observable
- state space
- finite state
- decision processes
- reinforcement learning
- dynamic programming
- reward function
- action space
- markov decision problems
- policy evaluation
- decision theoretic planning
- policy iteration algorithm
- reachability analysis
- total reward
- stationary policies
- transition matrices
- long run
- lower bound
- decision problems
- upper and lower bounds
- planning under uncertainty
- partially observable markov decision processes
- continuous state spaces
- expected reward
- factored mdps
- state dependent
- action sets
- learning algorithm
- semi markov decision processes
- model based reinforcement learning
- reinforcement learning algorithms
- real time dynamic programming
- sufficient conditions
- control policies
- initial state
- multistage