A Non-Negative Factorization approach to node pooling in Graph Convolutional Neural Networks
This addresses the need for efficient graph coarsening in machine learning applications, but it appears incremental as it builds on existing NMF and graph convolution methods.
The paper tackles the problem of subsampling in graph-structured data by introducing a pooling mechanism based on Non-Negative Matrix Factorization (NMF) for Graph Convolutional Neural Networks, resulting in significant improvements in predictive performance on graph classification benchmarks compared to non-pooled models.
The paper discusses a pooling mechanism to induce subsampling in graph structured data and introduces it as a component of a graph convolutional neural network. The pooling mechanism builds on the Non-Negative Matrix Factorization (NMF) of a matrix representing node adjacency and node similarity as adaptively obtained through the vertices embedding learned by the model. Such mechanism is applied to obtain an incrementally coarser graph where nodes are adaptively pooled into communities based on the outcomes of the non-negative factorization. The empirical analysis on graph classification benchmarks shows how such coarsening process yields significant improvements in the predictive performance of the model with respect to its non-pooled counterpart.