Combinatorial Correlation Clustering

Correlation Clustering is a classic clustering objective arising in numerous machine learning and data mining applications. Given a graph G= (V, E), the goal is to partition the vertex set into clusters so as to minimize the number of edges between clusters plus the number of edges missing within clusters. The problem is APX-hard and the best known polynomial time approximation factor is 1. 73 by Cohen-Addad, Lee, Li, and Newman FOCS'23. They use an LP with |V|^1/^{ (1) } variables for some small. However, due to the practical relevance of correlation clustering, there has also been great interest in getting more efficient sequential and parallel algorithms. The classic combinatorial pivot algorithm of Ailon, Charikar and Newman JACM'08 provides a 3-approximation in linear time. Like most other algorithms discussed here, this uses randomization. Recently, Behnezhad, Charikar, Ma and Tan FOCS'22 presented a 3+-approximate solution for solving problem in a constant number of rounds in the Massively Parallel Computation (MPC) setting. Very recently, Cao, Huang, Su SODA'24 provided a 2. 4-approximation in a polylogarithmic number of rounds in the MPC model and in O (|E|^1. 5) time in the classic sequential setting. They asked whether it is possible to get a better than 3-approximation in near-linear time? We resolve this problem with an efficient combinatorial algorithm providing a drastically better approximation factor. It achieves a 2-2/13 < 1. 847-approximation in sub-linear (O (|V|) ) sequential time or in sub-linear (O (|V|) ) space in the streaming setting. In the MPC model, we give an algorithm using only a constant number of rounds that achieves a 2-1/8 < 1. 876-approximation.