Deep Exploration by Planning With Uncertainty in Deep Model Based Reinforcement Learning

Deep, model based reinforcement learning has shown state of the art, human-exceeding performance in many challenging domains. <br/>Low sample efficiency and limited exploration remain however as leading obstacles in the field. <br/>In this work, we incorporate epistemic uncertainty into planning for better exploration.<br/>We develop a low-cost...

Action Selection Policies for Walking Monte Carlo Tree Search

Recent Reinforcement Learning methods have combined function approximation and Monte Carlo Tree Search and are able to learn by self-play up to a very high level in several games such as Go and Hex. One aspect in this combination<br/>that has not had a lot of attention is the action selection policy during self-play, which could influence the...

Smart Start: A Directed and Persistent Exploration Framework for Reinforcement Learning

An important problem in reinforcement learning is the exploration-exploitation dilemma. Especially for environments with sparse or misleading rewards it has proven difficult to construct a good exploration strategy. For discrete domains good exploration strategies have been devised, but are often nontrivial to implement on more complex domains...