"Norm spaces rooted in neural networks and their applications"Shuai LuWe revisit several neural network-derived norm spaces, encompassing (extended) Barron spaces, variation spaces, Radon-BV spaces, and spectral Barron spaces. We systematically investigate the properties of these spaces and explore their applications within regularization schemes and inverse problems in partial differential equations. It is joint work with Yuanyuan Li (Fudan), Peter Mathé (WIAS) and Sergei Pereverzev (RICAM). |
« back