Local2Global: A distributed approach for scaling representation learning on graphs

TL;DR

The paper introduces a decentralized 'local2global' method for scalable graph representation learning that independently trains local embeddings on subgraphs and then aligns them globally, enabling large-scale applications.

Contribution

It presents a novel local2global approach that avoids costly parameter synchronization, allowing scalable and efficient graph embedding on large, distributed datasets.

Findings

Achieves a good trade-off between scale and accuracy in edge reconstruction.

Effective in semi-supervised classification tasks.

Useful for anomaly detection in cybersecurity networks.

Abstract

We propose a decentralised "local2global"' approach to graph representation learning, that one can a-priori use to scale any embedding technique. Our local2global approach proceeds by first dividing the input graph into overlapping subgraphs (or "patches") and training local representations for each patch independently. In a second step, we combine the local representations into a globally consistent representation by estimating the set of rigid motions that best align the local representations using information from the patch overlaps, via group synchronization. A key distinguishing feature of local2global relative to existing work is that patches are trained independently without the need for the often costly parameter synchronization during distributed training. This allows local2global to scale to large-scale industrial applications, where the input graph may not even fit into memory and may be stored in a distributed manner. We apply local2global on data sets of different sizes and show that our approach achieves a good trade-off between scale and accuracy on edge reconstruction and semi-supervised classification. We also consider the downstream task of anomaly detection and show how one can use local2global to highlight anomalies in cybersecurity networks.