Fast Distributed Approximation for Max-Cut

TL;DR

This paper introduces near-optimal randomized and deterministic distributed algorithms for Max-Cut and Max-Dicut problems, achieving high approximation ratios efficiently in the classic distributed models.

Contribution

It presents the first almost optimal distributed algorithms for Max-Cut in the LOCAL and CONGEST models, including a deterministic 1/3-approximation for Max-Dicut, improving previous bounds.

Findings

Achieves (1-ε)-approximation for Max-Cut on bipartite graphs in CONGEST.

Provides a 1/3-approximation for Max-Dicut in distributed models.

Uses a greedy approach for submodular maximization, of independent interest.

Abstract

Finding a maximum cut is a fundamental task in many computational settings. Surprisingly, it has been insufficiently studied in the classic distributed settings, where vertices communicate by synchronously sending messages to their neighbors according to the underlying graph, known as the $\mathcal{LOCAL}$ or $\mathcal{CONGEST}$ models. We amend this by obtaining almost optimal algorithms for Max-Cut on a wide class of graphs in these models. In particular, for any $\epsilon > 0$, we develop randomized approximation algorithms achieving a ratio of $(1-\epsilon)$ to the optimum for Max-Cut on bipartite graphs in the $\mathcal{CONGEST}$ model, and on general graphs in the $\mathcal{LOCAL}$ model. We further present efficient deterministic algorithms, including a $1/3$-approximation for Max-Dicut in our models, thus improving the best known (randomized) ratio of $1/4$. Our algorithms make non-trivial use of the greedy approach of Buchbinder et al. (SIAM Journal on Computing, 2015) for maximizing an unconstrained (non-monotone) submodular function, which may be of independent interest.