Nov 15, 2024: MACHINE LEARNING FOR STREAMFLOW FORECASTING: A PRIMER FOR THE PRACTITIONER
Date/Time: Friday, November 15th, 2024, 1 pm to 4 pm EST
About the Event:
This short workshop aims to demonstrate core machine learning (ML) concepts and show practitioners how they can be used for streamflow forecasting. The workshop will use interactive Colab notebooks to guide participants through the ML streamflow forecasting “pipeline”. The goal is for participants to obtain relevant experience and practical ML tools that can be translated into their practice.
Long Short-Term Memory networks (LSTMs) will be introduced and used to generate streamflow forecasts using open-source datasets. Topics will include data preprocessing, model training, and out-of-sample verification.
Facilitators: John Quilty and Mohammad Sina Jahangir
Bio – John Quilty: Dr. John Quilty is an Assistant Professor in the Department of Civil and Environmental Engineering at the University of Waterloo and the Sinnathamby Professor in AI for Sustainable Solutions. His research focuses on innovative tools for addressing the issues of nonlinearity, multiscale change, and uncertainty in hydrological forecasting. Prof. Quilty also serves as an Associate Editor for Journal of Hydrology.
Bio – Mohammad Sina Jahangir: Dr. Mohammad Sina Jahangir is a research associate in the Department of Bioresource Engineering at McGill University and an alumnus of the University of Waterloo. He specializes in the development, design, optimization, and deployment of deep learning models. His research is primarily centered on developing probabilistic and generative deep learning models for multiscale forecasting.
Prerequisites: A laptop computer that can connect wirelessly to the internet. Gmail account for accessing Google Collab/Drive.