Büchi Objectives in Countable MDPs.
Stefan KieferRichard MayrMahsa ShirmohammadiPatrick TotzkePublished in: ICALP (2019)
Keyphrases
- markov decision processes
- average cost
- reinforcement learning
- state space
- markov chain
- factored mdps
- multiple objectives
- finite horizon
- finite number
- factored markov decision processes
- neural network
- model based reinforcement learning
- state and action spaces
- probabilistic planning
- markov decision process
- long run
- optimal policy
- dynamic programming
- feature selection
- partially observable
- initial state
- information gain
- linear program
- linear programming
- search space
- stochastic shortest path