Alternating Good-for-MDP Automata.
Ernst Moritz HahnMateo PerezSven ScheweFabio SomenziAshutosh TrivediDominik WojtczakPublished in: CoRR (2022)
Keyphrases
- finite state
- markov decision processes
- markov decision process
- optimal policy
- tree automata
- state space
- cellular automata
- reinforcement learning
- finite automata
- markov chain
- probabilistic automata
- action sets
- linear program
- planning under uncertainty
- policy iteration
- reward function
- linear programming
- dynamic programming algorithms
- utility function
- dynamic programming
- markov decision problems
- automata theoretic
- factored mdps
- finite automaton
- lattice gas
- database
- decision theoretic planning
- action space
- model checking
- natural language
- learning algorithm
- neural network