English

A common variable minimax theorem for graphs

Spectral Theory 2022-03-03 v1 Machine Learning

Abstract

Let G={G1=(V,E1),,Gm=(V,Em)}\mathcal{G} = \{G_1 = (V, E_1), \dots, G_m = (V, E_m)\} be a collection of mm graphs defined on a common set of vertices VV but with different edge sets E1,,EmE_1, \dots, E_m. Informally, a function f:VRf :V \rightarrow \mathbb{R} is smooth with respect to Gk=(V,Ek)G_k = (V,E_k) if f(u)f(v)f(u) \sim f(v) whenever (u,v)Ek(u, v) \in E_k. We study the problem of understanding whether there exists a nonconstant function that is smooth with respect to all graphs in G\mathcal{G}, simultaneously, and how to find it if it exists.

Keywords

Cite

@article{arxiv.2107.14747,
  title  = {A common variable minimax theorem for graphs},
  author = {Ronald R. Coifman and Nicholas F. Marshall and Stefan Steinerberger},
  journal= {arXiv preprint arXiv:2107.14747},
  year   = {2022}
}

Comments

21 pages, 11 figures

R2 v1 2026-06-24T04:41:48.118Z