Towards Instance-Optimality in Online PAC Reinforcement Learning

Several recent works have proposed instance-dependent upper bounds on the number of episodes needed to identify, with probability $1-\delta$, an $\varepsilon$-optimal policy in finite-horizon tabular Markov Decision Processes (MDPs). These upper bounds feature various complexity measures for the MDP...

Ausführliche Beschreibung

Bibliographische Detailangaben

Hauptverfasser:	Al-Marjani, Aymen, Tirinzoni, Andrea, Kaufmann, Emilie
Format:	Artikel
Sprache:	eng
Schlagworte:	Computer Science - Learning Statistics - Machine Learning
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!