Cheeger Inequalities for Submodular Transformations

TL;DR

This paper generalizes Cheeger inequalities to submodular transformations, unifying and extending spectral graph theory to broader classes of graphs, hypergraphs, and information measures, with approximation algorithms for eigenvalue computation.

Contribution

It introduces a new framework for Cheeger inequalities for submodular transformations, encompassing graphs, hypergraphs, and information measures, and provides approximation algorithms for eigenvalue computation.

Findings

Unified Cheeger inequality for submodular transformations

Polynomial-time $O(\log n)$-approximation algorithm for symmetric case

Polynomial-time $O(\log^2 n + \log n imes \log m)$-approximation for general case

Abstract

The Cheeger inequality for undirected graphs, which relates the conductance of an undirected graph and the second smallest eigenvalue of its normalized Laplacian, is a cornerstone of spectral graph theory. The Cheeger inequality has been extended to directed graphs and hypergraphs using normalized Laplacians for those, that are no longer linear but piecewise linear transformations. In this paper, we introduce the notion of a submodular transformation $F:\{0,1\}^n \to \mathbb{R}^m$, which applies $m$ submodular functions to the $n$-dimensional input vector, and then introduce the notions of its Laplacian and normalized Laplacian. With these notions, we unify and generalize the existing Cheeger inequalities by showing a Cheeger inequality for submodular transformations, which relates the conductance of a submodular transformation and the smallest non-trivial eigenvalue of its normalized Laplacian. This result recovers the Cheeger inequalities for undirected graphs, directed graphs, and hypergraphs, and derives novel Cheeger inequalities for mutual information and directed information. Computing the smallest non-trivial eigenvalue of a normalized Laplacian of a submodular transformation is NP-hard under the small set expansion hypothesis. In this paper, we present a polynomial-time $O(\log n)$-approximation algorithm for the symmetric case, which is tight, and a polynomial-time $O(\log^2n+\log n \cdot \log m)$-approximation algorithm for the general case. We expect the algebra concerned with submodular transformations, or \emph{submodular algebra}, to be useful in the future not only for generalizing spectral graph theory but also for analyzing other problems that involve piecewise linear transformations, e.g., deep learning.