6224912 – Deep Learning in Hydrological Modeling

This is an interdisciplinary course that aims to provide students (Master and PhD students) with a solid foundation in applying advanced deep learning techniques to tackle complex environmental modeling problems, specifically supervised learning of time series prediction problems. The course will explore the role of deep learning in simulating, predicting, and understanding environmental processes and systems. Core learning outcomes will be the ability to devlop and implement state of the art deep learning models, such as Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks, to analyze large-scale environmental data sets. Students witll gain practical experience in designing and training deep learning models using the PyTorch framework and gain deep understanding of the underlying principles and algorithms, such as backpropagation, optimization techniques, and regularization. In addition to hands-on programming assignments, the course will involve critical analysis of relevant research papers and case studies.