Projection by Convolution: Optimal Sample Complexity for Reinforcement Learning in Continuous-Space MDPs.

Davide Maran, Alberto Maria Metelli, Matteo Papini, Marcello Restelli

Browse the full COLT paper archive.