Approximation of Generalized Ridge Functions in High Dimensions – Sandra Keiper

Explore the mathematical foundations of approximating high-dimensional functions in this 47-minute lecture by Sandra Keiper from the Alan Turing Institute. Delve into the challenges of reconstructing complex processes with numerous parameters, focusing on overcoming the curse of dimensionality. Learn about modern approaches that make structural assumptions to bypass limitations, such as low intrinsic dimensionality, partial separability, and sparse representations. Gain insights into the rich theory developed over the past decade for various interesting models in multivariate approximation theory, high-dimensional integration, and non-parametric regression.