On Overfitting and Asymptotic Bias in Batch Reinforcement Learning with Partial Observability.
Vincent François-LavetGuillaume RabusseauJoelle PineauDamien ErnstRaphael FonteneauPublished in: J. Artif. Intell. Res. (2019)
Keyphrases
- partial observability
- reinforcement learning
- partially observable
- symbolic model checking
- planning problems
- fully observable
- markov decision process
- markov decision processes
- belief state
- learning agent
- belief space
- state space
- supervised learning
- model free
- partially observable markov decision processes
- planning under partial observability
- function approximation
- partial information
- learning capabilities
- learning algorithm