Geometric Scattering Attention Networks

TL;DR

This paper introduces a novel geometric scattering attention network (GSAN) that adaptively combines scattering and GCN features for improved semi-supervised node classification, while providing insights into spectral information through attention weights.

Contribution

The paper proposes an attention-based architecture that learns node-specific weights to combine scattering and GCN channels, enhancing representation learning.

Findings

GSAN outperforms previous models in semi-supervised node classification

The model enables spectral analysis of node features via attention weights

Adaptive learning of node-wise weights improves feature integration

Abstract

Geometric scattering has recently gained recognition in graph representation learning, and recent work has shown that integrating scattering features in graph convolution networks (GCNs) can alleviate the typical oversmoothing of features in node representation learning. However, scattering often relies on handcrafted design, requiring careful selection of frequency bands via a cascade of wavelet transforms, as well as an effective weight sharing scheme to combine low- and band-pass information. Here, we introduce a new attention-based architecture to produce adaptive task-driven node representations by implicitly learning node-wise weights for combining multiple scattering and GCN channels in the network. We show the resulting geometric scattering attention network (GSAN) outperforms previous networks in semi-supervised node classification, while also enabling a spectral study of extracted information by examining node-wise attention weights.