Parity objectives in countable MDPs.
Stefan KieferRichard MayrMahsa ShirmohammadiDominik WojtczakPublished in: LICS (2017)
Keyphrases
- markov decision processes
- average cost
- reinforcement learning
- state space
- error correction
- factored mdps
- finite state
- multiple objectives
- real time dynamic programming
- planning under uncertainty
- policy iteration
- markov decision process
- optimal policy
- markov chain
- initial state
- finite horizon
- decision processes
- dynamic programming
- partially observable
- infinite horizon
- situation calculus
- probabilistic planning
- action sets
- learning algorithm