Explore the principles of Geometric Deep Learning in this comprehensive tutorial focusing on Graph Neural Networks (GNNs). Delve into the anatomy of GNNs and understand the motivations behind various architectural choices. Discover the connection between GNNs and differential equations on graphs, gaining insights through the lenses of differential geometry and algebraic topology. Learn from Michael Bronstein of the University of Oxford as he presents this in-depth talk at the 2022 SIAM Conference on Mathematics of Data Science, offering a unified perspective on neural network architectures through symmetry and invariance principles.