Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactions

Proceedings of the AAAI Conference on Artificial Intelligence, vol. 37, pp. 6779–6787

Abstract

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 an 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.

Authors 5

1. Anson Bastos

   Indian Institute of Technology Hyderabad

   Affiliation as printed

   Indian Institute of Technology Hyderabad

2. Abhishek Nadgeri Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen

3. Kuldeep Singh
   Affiliation as printed

   Cerence GmbH

4. Toyotaro Suzumura

   IBM Research - Thomas J. Watson Research Center

   Affiliation as printed

   IBM T.J. Watson Research Center

5. Manish Singh

   Indian Institute of Technology Hyderabad

   Affiliation as printed

   Indian Institute of Technology, Hyderabad