Sampling-based Distributed Training with Message Passing Neural Network

arXiv:2402.15106v1 Announce Type: new
Abstract: In this study, we introduce a domain-decomposition-based distributed training and inference approach for message-passing neural networks (MPNN). Our objective is to address the challenge of scaling edge-based graph neural networks as the number of nodes increases. Through our distributed training approach, coupled with Nystr\"om-approximation sampling techniques, we present a scalable graph neural network, referred to as DS-MPNN (D and S standing for distributed and sampled, respectively), capable of scaling up to $O(10^5)$ nodes. We validate …

abstract approximation arxiv challenge cs.dc cs.lg distributed domain edge graph graph neural networks inference network networks neural network neural networks nodes physics.flu-dyn sampling scaling study through training type