HDR Defense of Tanujit Chakraborty 

"Contributions to Time Series Forecasting and Statistical Learning"

Abstract: This dissertation covers the work conducted by the author, mostly as an Assistant Professor (2021–2024) and Associate Professor (2024 onwards) at the Sorbonne Université Abu Dhabi. During this period, the author strengthened his contributions to (spatio)temporal forecasting and statistical learning, including imbalanced learning, online learning, and learning patterns in network data. Motivated by challenges arising in climate, health, and economic domains, this HDR thesis presents theoretical and methodological contributions to time series forecasting and statistical learning, integrating data-driven modeling with statistical rigor. This manuscript does not aim to present an exhaustive description of the results developed by
the author, but rather a synthetic view of his main contributions.