Adaptive Stacked Graph Filter

Hoang NT, Takanori Maehara, Tsuyoshi Murata

Keywords: Graph Convolutional Network, vertex classification, graph signal processing, adaptive graph filter

Abstract: We study Graph Convolutional Networks (GCN) from the graph signal processing viewpoint by addressing a difference between learning graph filters with fully-connected weights versus trainable polynomial coefficients. We find that by stacking graph filters with learnable polynomial parameters, we can build a highly adaptive and robust vertex classification model. Our treatment here relaxes the low-frequency (or equivalently, high homophily) assumptions in existing vertex classification models, resulting a more ubiquitous solution in terms of spectral properties. Empirically, by using only one hyper-parameter setting, our model achieves strong results on most benchmark datasets across the frequency spectrum.

One-sentence Summary: We show that simply learning the polynomial coefficients of a graph filter can lead to a highly adaptive semi-supervised vertex classification model.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=-Tc5mpYIG