Metrics and Isometries for Convex Functions
Ben Li, Fabian Mussnig

TL;DR
This paper introduces new metrics for convex functions that align with epi-convergence and classifies their isometries, advancing the understanding of functional geometry.
Contribution
It develops a class of metrics for convex functions, proves their equivalence to epi-convergence, and classifies their isometries, extending geometric analysis tools.
Findings
Metrics are equivalent to epi-convergence.
Full classification of isometries for these metrics.
Introduction of new Hausdorff-like metrics for convex functions.
Abstract
We introduce a class of functional analogs of the symmetric difference metric on the space of coercive convex functions on with full-dimensional domain. We show that convergence with respect to these metrics is equivalent to epi-convergence. Furthermore, we give a full classification of all isometries with respect to some of the new metrics. Moreover, we introduce two new functional analogs of the Hausdorff metric on the spaces of coercive convex functions and super-coercive convex functions, respectively, and prove equivalence to epi-convergence.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
