Formally Verified Solution Methods for Infinite-Horizon Markov Decision Processes.
Maximilian SchäffelerMohammad AbdulazizPublished in: CoRR (2022)
Keyphrases
- markov decision processes
- infinite horizon
- optimal policy
- finite horizon
- reinforcement learning
- policy iteration
- finite state
- state space
- dynamic programming
- average cost
- partially observable
- transition matrices
- optimal control
- markov decision process
- machine learning
- single item
- probabilistic planning
- decision problems
- optimal solution
- decision processes
- average reward