Successor Uncertainties: Exploration and Uncertainty in Temporal Difference Learning

Abstract

Posterior sampling for reinforcement learning (PSRL) is an effective method for balancing exploration and exploitation in reinforcement learning. Randomised value functions (RVF) can be viewed as a promising approach to scaling PSRL. However, we show that most contemporary algorithms combining RVF with neural network function approximation do not possess the properties which make PSRL effective, and provably fail in sparse reward problems. Moreover, we find that propagation of uncertainty, a property of PSRL previously thought important for exploration, does not preclude this failure. We use these insights to design Successor Uncertainties (SU), a cheap and easy to implement RVF algorithm that retains key properties of PSRL. SU is highly effective on hard tabular exploration benchmarks. Furthermore, on the Atari 2600 domain, it surpasses human performance on 38 of 49 games tested (achieving a median human normalised score of 2.09), and outperforms its closest RVF competitor, Bootstrapped DQN, on 36 of those.

arXiv commentCamera ready version, NeurIPS 2019

Cited by in corpus (10)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertainty
• Actionable Models: Unsupervised Offline Reinforcement Learning of Robotic Skills
• Kalman meets Bellman: Improving Policy Evaluation through Value Tracking
• Temporal Difference Uncertainties as a Signal for Exploration
• Uncertainty-sensitive Learning and Planning with Ensembles
• A Survey of Exploration Methods in Reinforcement Learning
• PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learning
• Temporally-Extended ε-Greedy Exploration
• Implications of Human Irrationality for Reinforcement Learning