A Gaussian Belief Propagation Solver for Large Scale Support Vector Machines

TL;DR

This paper presents a scalable, parallel Gaussian Belief Propagation algorithm for large-scale support vector machines, achieving high accuracy and speed on supercomputers with thousands of nodes.

Contribution

It introduces the largest parallel implementation of belief propagation for SVMs, enabling efficient distributed training on massive datasets.

Findings

Comparable accuracy to existing SVM solvers

Significantly faster for large datasets

Scalable to over 1,000 computing nodes

Abstract

Support vector machines (SVMs) are an extremely successful type of classification and regression algorithms. Building an SVM entails solving a constrained convex quadratic programming problem, which is quadratic in the number of training samples. We introduce an efficient parallel implementation of an support vector regression solver, based on the Gaussian Belief Propagation algorithm (GaBP). In this paper, we demonstrate that methods from the complex system domain could be utilized for performing efficient distributed computation. We compare the proposed algorithm to previously proposed distributed and single-node SVM solvers. Our comparison shows that the proposed algorithm is just as accurate as these solvers, while being significantly faster, especially for large datasets. We demonstrate scalability of the proposed algorithm to up to 1,024 computing nodes and hundreds of thousands of data points using an IBM Blue Gene supercomputer. As far as we know, our work is the largest parallel implementation of belief propagation ever done, demonstrating the applicability of this algorithm for large scale distributed computing systems.