Alternating Good-for-MDPs Automata.
Ernst Moritz HahnMateo PerezSven ScheweFabio SomenziAshutosh TrivediDominik WojtczakPublished in: ATVA (2022)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- factored mdps
- average cost
- state space
- finite state machines
- finite automata
- policy iteration
- cellular automata
- tree automata
- regular expressions
- markov decision problems
- optimal policy
- reward function
- finite horizon
- decision theoretic planning
- dynamic programming
- markov decision process
- average reward
- planning under uncertainty
- probabilistic automata
- lattice gas
- probabilistic planning
- semi markov decision processes
- partially observable markov decision processes
- multiple agents
- turing machine
- finite state automata
- temporal logic
- linear programming
- finite automaton
- model based reinforcement learning
- real time dynamic programming