Adaptive Optimization of Isogeometric Multi-Patch Discretizations Using Artificial Neural Networks

The core message of this article is to present a novel approach for adaptively optimizing the parameterization of isogeometric multi-patch discretizations using artificial neural networks. The method aims to find a suitable isogeometric function space that minimizes the approximation error for a given partial differential equation without sacrificing the tensor-product structure of the underlying spline space.