Ensemble Quantile Networks: Uncertainty-Aware Reinforcement Learning with Applications in Autonomous Driving
This addresses the need for uncertainty-aware decision-making in autonomous driving, though it is an incremental improvement by integrating existing techniques for uncertainty estimation.
The paper tackled the problem of uncertainty estimation in reinforcement learning for autonomous driving by introducing Ensemble Quantile Networks (EQN), which combine distributional RL and ensemble methods to estimate both aleatoric and epistemic uncertainty, enabling the agent to balance risk and time efficiency in occluded intersections and avoid dangerous decisions outside the training distribution.
Reinforcement learning (RL) can be used to create a decision-making agent for autonomous driving. However, previous approaches provide only black-box solutions, which do not offer information on how confident the agent is about its decisions. An estimate of both the aleatoric and epistemic uncertainty of the agent's decisions is fundamental for real-world applications of autonomous driving. Therefore, this paper introduces the Ensemble Quantile Networks (EQN) method, which combines distributional RL with an ensemble approach, to obtain a complete uncertainty estimate. The distribution over returns is estimated by learning its quantile function implicitly, which gives the aleatoric uncertainty, whereas an ensemble of agents is trained on bootstrapped data to provide a Bayesian estimation of the epistemic uncertainty. A criterion for classifying which decisions that have an unacceptable uncertainty is also introduced. The results show that the EQN method can balance risk and time efficiency in different occluded intersection scenarios, by considering the estimated aleatoric uncertainty. Furthermore, it is shown that the trained agent can use the epistemic uncertainty information to identify situations that the agent has not been trained for and thereby avoid making unfounded, potentially dangerous, decisions outside of the training distribution.