The Potential of the Return Distribution for Exploration in RL

06/11/2018

∙

by   Thomas M. Moerland, et al.

∙

0

∙

share

This paper studies the potential of the return distribution for exploration in deterministic environments. We study network losses and propagation mechanisms for Gaussian, Categorical and Mixture of Gaussian distributions. Combined with exploration policies that leverage this return distribution, we solve, for example, a randomized Chain task of length 100, which has not been reported before when learning with neural networks.