Adaptive Sampling Towards Fast Graph Representation Learning

TL;DR

This paper introduces an adaptive, layer-wise sampling method for Graph Convolutional Networks that improves scalability and training efficiency on large graphs, with enhanced message passing and reduced variance.

Contribution

It proposes a novel adaptive sampling strategy with fixed-size neighborhoods and skip connections to accelerate GCN training on large-scale graphs.

Findings

Faster convergence speed in training GCNs.

Improved classification accuracy on benchmark datasets.

Effective variance reduction in sampling process.

Abstract

Graph Convolutional Networks (GCNs) have become a crucial tool on learning representations of graph vertices. The main challenge of adapting GCNs on large-scale graphs is the scalability issue that it incurs heavy cost both in computation and memory due to the uncontrollable neighborhood expansion across layers. In this paper, we accelerate the training of GCNs through developing an adaptive layer-wise sampling method. By constructing the network layer by layer in a top-down passway, we sample the lower layer conditioned on the top one, where the sampled neighborhoods are shared by different parent nodes and the over expansion is avoided owing to the fixed-size sampling. More importantly, the proposed sampler is adaptive and applicable for explicit variance reduction, which in turn enhances the training of our method. Furthermore, we propose a novel and economical approach to promote the message passing over distant nodes by applying skip connections. Intensive experiments on several benchmarks verify the effectiveness of our method regarding the classification accuracy while enjoying faster convergence speed.