Good-for-MDPs Automata.
Ernst Moritz HahnMateo PerezFabio SomenziAshutosh TrivediSven ScheweDominik WojtczakPublished in: CoRR (2019)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- factored mdps
- state space
- optimal policy
- dynamic programming
- cellular automata
- finite automata
- probabilistic automata
- planning under uncertainty
- average cost
- finite horizon
- regular expressions
- markov decision problems
- tree automata
- finite state machines
- lattice gas
- finite state automata
- policy iteration
- decision theoretic planning
- average reward
- reward function
- policy evaluation
- decision diagrams
- dec pomdps
- factored markov decision processes
- real time dynamic programming
- infinite horizon
- machine learning
- state and action spaces
- multi agent
- initial state
- continuous state and action spaces