Know what it does not know: Improving Offline Deep Reinforcement Learning with Uncertainty Estimation

Offline reinforcement learning, or learning from a fixed data set, is an attractive alternative to online reinforcement learning. Offline reinforcement learning promises to address the cost and safety implications of taking numerous random or bad actions online, which is a crucial aspect of traditional reinforcement learning that makes it...

Operational flood forecasting, warning and response for multi-scale flood risks in developing cities

Flood early warning systems are recognized as one of the most effective flood risk management instruments when correctly embedded in comprehensive flood risk management strategies and policies. Many efforts around the world are being put in place to advance the components that determine the effectiveness of a flood early warning system. The aim...