Deterministic Distributed Edge-Coloring with Fewer Colors

TL;DR

This paper introduces new deterministic distributed algorithms for edge-coloring that use fewer colors than previous methods, achieving near-optimal color bounds in polylogarithmic time for graphs with certain degree conditions.

Contribution

It presents the first deterministic algorithms to color edges with fewer than 2Δ-1 colors in polylogarithmic time, improving upon prior work and solving a longstanding open problem.

Findings

Achieves $(1+o(1))elta$-edge-coloring in polylogarithmic time for large elta.

Provides a $3elta/2$-edge-coloring for smaller elta in polylogarithmic time.

Breaks the natural barrier of $2elta-1$ colors in deterministic distributed algorithms.

Abstract

We present a deterministic distributed algorithm, in the LOCAL model, that computes a $(1+o(1))\Delta$-edge-coloring in polylogarithmic-time, so long as the maximum degree $\Delta=\tilde{\Omega}(\log n)$. For smaller $\Delta$, we give a polylogarithmic-time $3\Delta/2$-edge-coloring. These are the first deterministic algorithms to go below the natural barrier of $2\Delta-1$ colors, and they improve significantly on the recent polylogarithmic-time $(2\Delta-1)(1+o(1))$-edge-coloring of Ghaffari and Su [SODA'17] and the $(2\Delta-1)$-edge-coloring of Fischer, Ghaffari, and Kuhn [FOCS'17], positively answering the main open question of the latter. The key technical ingredient of our algorithm is a simple and novel gradual packing of judiciously chosen near-maximum matchings, each of which becomes one of the color classes.