RAMA: A Rapid Multicut Algorithm on GPU

TL;DR

RAMA introduces a GPU-based parallel primal-dual algorithm for the multicut problem, achieving significant speedups and enabling large-scale graph clustering solutions in seconds.

Contribution

It presents a novel GPU-accelerated multicut algorithm with recursive cycle detection, message passing, and edge contraction, improving speed and scalability over traditional CPU methods.

Findings

Achieves 10-100x speedup over CPU algorithms.

Solves large-scale problems with up to 10^8 variables in seconds.

Maintains solution quality comparable to traditional methods.

Abstract

We propose a highly parallel primal-dual algorithm for the multicut (a.k.a. correlation clustering) problem, a classical graph clustering problem widely used in machine learning and computer vision. Our algorithm consists of three steps executed recursively: (1) Finding conflicted cycles that correspond to violated inequalities of the underlying multicut relaxation, (2) Performing message passing between the edges and cycles to optimize the Lagrange relaxation coming from the found violated cycles producing reduced costs and (3) Contracting edges with high reduced costs through matrix-matrix multiplications. Our algorithm produces primal solutions and lower bounds that estimate the distance to optimum. We implement our algorithm on GPUs and show resulting one to two orders-of-magnitudes improvements in execution speed without sacrificing solution quality compared to traditional sequential algorithms that run on CPUs. We can solve very large scale benchmark problems with up to $\mathcal{O}(10^8)$ variables in a few seconds with small primal-dual gaps. Our code is available at https://github.com/pawelswoboda/RAMA.