Synthesis of Pareto-optimal Policies for Continuous-Time Markov Decision Processes.
Naif AlasmariRadu CalinescuPublished in: SEAA (2022)
Keyphrases
- markov decision processes
- optimal policy
- state space
- reinforcement learning
- decision problems
- stationary policies
- dynamic programming
- finite state
- infinite horizon
- policy iteration
- finite horizon
- markov chain
- average reward
- state dependent
- multistage
- decision processes
- average cost
- dynamical systems
- long run
- decision theoretic planning
- sufficient conditions
- partially observable
- markov decision process
- reward function
- optimal control
- initial state
- reinforcement learning algorithms
- total reward
- action space
- discounted reward
- markov decision problems
- heuristic search
- semi markov decision processes
- belief state
- planning problems