The predictive power and overall computational efficiency of Diffusion-convolutional neural networks make them an attractive choice for node classification tasks. However, a naive dense-tensor-based implementation of DCNNs leads to O(N2) memory complexity which is prohibitive for large graphs. In this paper, we introduce a simple method for thresholding input graphs that provably reduces memory requirements of DCNNs to O(N) (i.e. linear in the number of nodes in the input) without significantly affecting predictive performance.
@article{arxiv.1710.09813,
title = {Sparse Diffusion-Convolutional Neural Networks},
author = {James Atwood and Siddharth Pal and Don Towsley and Ananthram Swami},
journal= {arXiv preprint arXiv:1710.09813},
year = {2017}
}