Uncertainty-Based Out-of-Distribution Detection in Deep Reinforcement Learning
This addresses the challenge of identifying unfamiliar inputs for reinforcement learning agents, though it appears incremental as it builds on existing uncertainty methods in a specific domain.
The paper tackled the problem of detecting out-of-distribution samples in deep reinforcement learning by applying uncertainty estimation techniques to value networks, finding that bootstrap-based approaches produced more reliable uncertainty estimates than dropout-based methods during training.
We consider the problem of detecting out-of-distribution (OOD) samples in deep reinforcement learning. In a value based reinforcement learning setting, we propose to use uncertainty estimation techniques directly on the agent's value estimating neural network to detect OOD samples. The focus of our work lies in analyzing the suitability of approximate Bayesian inference methods and related ensembling techniques that generate uncertainty estimates. Although prior work has shown that dropout-based variational inference techniques and bootstrap-based approaches can be used to model epistemic uncertainty, the suitability for detecting OOD samples in deep reinforcement learning remains an open question. Our results show that uncertainty estimation can be used to differentiate in- from out-of-distribution samples. Over the complete training process of the reinforcement learning agents, bootstrap-based approaches tend to produce more reliable epistemic uncertainty estimates, when compared to dropout-based approaches.