Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactions

Learning on evolving(dynamic) graphs has caught the attention of researchers as static methods exhibit limited performance in this setting. The existing for dynamic learn spatial features by local neighborhood aggregation, which essentially only captures low pass signals and interactions. In work, we go beyond current approaches to incorporate global effectively learning representations a dynamically evolving graph. We propose do so capturing spectrum Since graph would not consider history evolution evolves with time, an approach wavelets capture spectra. Further, framework that integrates captured spectra form these learnable into incorporating Experiments eight standard datasets show our method significantly outperforms related various tasks graphs.