On June 29th 2023 at 17:00 in the FMI Conference Hall meeting of the FMI Colloquium will be held.
Prof. Gergana Petrova, Texas A&M University, USA, is going to give a research presentation on topic: Lower bounds for Neural Network Approximation
We introduce a measure, called Lipschitz widths, of the optimal performance possible of certain nonlinear methods of approximation. They provide a theoretical benchmark for the approximation quality achieved by neural networks. We investigate these widths and prove Carl's type inequalities for the error of approximation of compact sets K by deep and shallow neural networks.
Prior to the lecture, colloquium participants are invited at 16:30 in the FMI Conference Hall for an informal discussion over coffee/tea.
You are welcome!