Speaker
Eloi Martinet
(JMU Würzburg)
Description
We propose a single-layer neural parametrization of convex sets by learning sublinear (positively homogeneous and convex) functions. Our networks explicitly represent both the support and gauge functions of a convex body. We prove a universal approximation theorem for convex sets under this parametrization. Empirically, we demonstrate the method on shape optimization and inverse design tasks, achieving accurate reconstruction of target shapes.
Author
Eloi Martinet
(JMU Würzburg)