In this manuscript we show for the first time how to train a single LSTM-based neural network as general hydrology model for hundreds of basins. Furthermore, we proposed the Entity-Aware LSTM (EA-LSTM) in which static features are used explicitly to subset the model for a specific entity (here a catchment).
Posts
PICO presentation at the EGU General Assembly 2019 on prediction in ungauged basins using LSTM based models.
Poster presentation at the EGU General Assembly 2019 on uncertainty estimation using MC-Dropout and LSTMs.
Video presentation in CUAHSI’s 2019 Spring Cyberseminar Series on Recent advances in big data machine learning in Hydrology.
Book chapter in the Explainable AI: Interpreting, Explaining and Visualizing Deep Learning (Editors Wojciech SamekGrégoire MontavonAndrea VedaldiLars Kai HansenKlaus-Robert Müller).
Presentation at the AGU 2018 Fall Meeting on experiments regarding the interpretability of LSTM states.
NeurIPS 2018 workshop paper, showing first results on using LSTMs for prediction in ungauged basins.