GIST: Distributed Training for Large-Scale Graph Convolutional Networks
This addresses the problem of efficiently training overparameterized GCNs on large graphs for researchers and practitioners in graph machine learning, representing a novel method rather than an incremental improvement.
The paper tackles the challenge of scaling training for large-scale graph convolutional networks (GCNs), particularly ultra-wide models, by proposing GIST, a distributed training framework that partitions parameters into sub-GCNs for parallel training, achieving SOTA performance on the Amazon2M dataset with a 32,768-dimensional model.
The graph convolutional network (GCN) is a go-to solution for machine learning on graphs, but its training is notoriously difficult to scale both in terms of graph size and the number of model parameters. Although some work has explored training on large-scale graphs (e.g., GraphSAGE, ClusterGCN, etc.), we pioneer efficient training of large-scale GCN models (i.e., ultra-wide, overparameterized models) with the proposal of a novel, distributed training framework. Our proposed training methodology, called GIST, disjointly partitions the parameters of a GCN model into several, smaller sub-GCNs that are trained independently and in parallel. In addition to being compatible with all GCN architectures and existing sampling techniques for efficient GCN training, GIST i) improves model performance, ii) scales to training on arbitrarily large graphs, iii) decreases wall-clock training time, and iv) enables the training of markedly overparameterized GCN models. Remarkably, with GIST, we train an astonishgly-wide 32,768-dimensional GraphSAGE model, which exceeds the capacity of a single GPU by a factor of 8x, to SOTA performance on the Amazon2M dataset.