Exploration by Distributional Reinforcement Learning
Machine Learning
2018-06-22 v2 Artificial Intelligence
Machine Learning
Abstract
We propose a framework based on distributional reinforcement learning and recent attempts to combine Bayesian parameter updates with deep reinforcement learning. We show that our proposed framework conceptually unifies multiple previous methods in exploration. We also derive a practical algorithm that achieves efficient exploration on challenging control tasks.
Cite
@article{arxiv.1805.01907,
title = {Exploration by Distributional Reinforcement Learning},
author = {Yunhao Tang and Shipra Agrawal},
journal= {arXiv preprint arXiv:1805.01907},
year = {2018}
}
Comments
IJCAI 2018