PhD Oral Defense: An ensemble wavelet-based stochastic data-driven framework for addressing nonlinearity, multiscale change, and uncertainty in water resources forecasting

PhD Oral Defense of John Quilty, Bioresource Engineering.

Data-driven forecasting (i.e., regression, machine learning, artificial intelligence, etc.) has become a popular and very useful alternative to physically-based and conceptual forecasting approaches in the water resources domain since such methods solely rely on statistical relationships between explanatory variables and the target process, require no explicit physical knowledge of the processes under study, are rapid to develop, have low-costs, and are easy to implement in real-time.  However, similar to physically-based and conceptual forecasting approaches, the nonlinear, multiscale, and uncertain nature of water resources provide challenges in the development of accurate and reliable data-driven forecasts.

Everyone in the McGill community is welcome to attend a PhD oral defense. Please join us in celebrating the accomplishments of our PhD candidates.