Universal Representation of Generalized Convex Functions and their Gradients
Fuente:
arXiv
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916005821808640 |
|---|---|
| author | Nehzati, Moeen |
| author_facet | Nehzati, Moeen |
| contents | A wide range of optimization problems can often be written in terms of generalized convex functions (GCFs). When this structure is present, it can convert certain nested bilevel objectives into single-level problems amenable to standard first-order optimization methods. We provide a new differentiable layer with a convex parameter space and show (Theorems 5.1 and 5.2) that it and its gradient are universal approximators for GCFs and their gradients. We demonstrate how this parameterization can be leveraged in practice by (i) learning optimal transport maps with general cost functions and (ii) learning optimal auctions of multiple goods. In both these cases, we show how our layer can be used to convert the existing bilevel or min-max formulations into single-level problems that can be solved efficiently with first-order methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_04477 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Universal Representation of Generalized Convex Functions and their Gradients Nehzati, Moeen Optimization and Control Machine Learning 91-08, 91-10, 91B68, 62P20, 90C26, 90C30, 65D40, 65K10, 49J52, 41A30 G.1.2; G.1.6; G.1.10; I.5.1 A wide range of optimization problems can often be written in terms of generalized convex functions (GCFs). When this structure is present, it can convert certain nested bilevel objectives into single-level problems amenable to standard first-order optimization methods. We provide a new differentiable layer with a convex parameter space and show (Theorems 5.1 and 5.2) that it and its gradient are universal approximators for GCFs and their gradients. We demonstrate how this parameterization can be leveraged in practice by (i) learning optimal transport maps with general cost functions and (ii) learning optimal auctions of multiple goods. In both these cases, we show how our layer can be used to convert the existing bilevel or min-max formulations into single-level problems that can be solved efficiently with first-order methods. |
| title | Universal Representation of Generalized Convex Functions and their Gradients |
| topic | Optimization and Control Machine Learning 91-08, 91-10, 91B68, 62P20, 90C26, 90C30, 65D40, 65K10, 49J52, 41A30 G.1.2; G.1.6; G.1.10; I.5.1 |
| url | https://arxiv.org/abs/2509.04477 |