Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactions
This work addresses the challenge of capturing global interactions in dynamic graphs for researchers and practitioners in graph learning, offering a novel integration of spectral and spatial features.
The paper tackled the problem of limited performance in dynamic graph learning by proposing a method to incorporate global features through learnable spectral wavelets, which capture evolving graph spectra and integrate them with spatial features. Experiments on eight datasets showed significant outperformance over related methods on various tasks.
Learning on evolving(dynamic) graphs has caught the attention of researchers as static methods exhibit limited performance in this setting. The existing methods for dynamic graphs learn spatial features by local neighborhood aggregation, which essentially only captures the low pass signals and local interactions. In this work, we go beyond current approaches to incorporate global features for effectively learning representations of a dynamically evolving graph. We propose to do so by capturing the spectrum of the dynamic graph. Since static methods to learn the graph spectrum would not consider the history of the evolution of the spectrum as the graph evolves with time, we propose a novel approach to learn the graph wavelets to capture this evolving spectra. Further, we propose a framework that integrates the dynamically captured spectra in the form of these learnable wavelets into spatial features for incorporating local and global interactions. Experiments on eight standard datasets show that our method significantly outperforms related methods on various tasks for dynamic graphs.