7–11 Sept 2026
Humboldt Universität zu Berlin
Europe/Berlin timezone

Parametrizing Convex Sets Using A Sublinear Layer

8 Sept 2026, 11:00
30m
DOR24/Floor 1-Room 103 - Lecture Hall (HU (Hegelplatz))

DOR24/Floor 1-Room 103 - Lecture Hall

HU (Hegelplatz)

HU Berlin Dorotheenstrasse 24 (Hegelplatz) 10117 Berlin
80
Show room on map
Free Boundary Problems in Data Science and Machine Learning Free Boundary Problems in Data Science and Machine Learning

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)

Presentation materials

There are no materials yet.