This course offers an in-depth look at some fundamental mathematical concepts driving recent advances in machine learning. Core topics encompass the theory of deep learning, large-scale and distributed optimization, causal inference, fairness, and safety in AI. Each topic will be explored through rigorous mathematical development complemented bypractical Python-based experiments. For the final assessment, students must present a topic of their choice from a provided list and resources, following the same format of rigorous mathematical exploration and Python experimentation.
This course reviews methods for machine learning based on the theory of optimal transportation. Two main algorithms will be studied: linear programming and Sinkhorn algorithm.