Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactions
Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactions
复制标题
DOI：
10.48550/arxiv.2211.11979
复制
发表时间：
2022-11
期刊：
影响因子：
--
通讯作者：
Anson Bastos;Abhishek Nadgeri;Kuldeep Singh;T. Suzumura;Manish Singh
中科院分区：
文献类型：
--
作者：
Anson Bastos;Abhishek Nadgeri;Kuldeep Singh;T. Suzumura;Manish Singh
进化（动态）图的学习引起了研究人员的注意，因为静态方法在这种情况下表现出有限的性能。现有的动态图方法通过局部邻域聚合来学习空间特征，本质上仅捕获低通信号和局部交互。在这项工作中，我们超越了当前的方法，结合全局特征来有效地学习动态演化图的表示。我们建议通过捕获动态图的频谱来实现这一点。由于学习图谱的静态方法不会考虑图谱随时间演变的演变历史，因此我们提出了一种学习图小波的方法来捕获这种演变的谱。此外，我们提出了一个框架，将这些可学习小波形式的动态捕获光谱集成到空间特征中，以合并局部和全局交互。在八个标准数据集上的实验表明，我们的方法在动态图的各种任务上显着优于相关方法。
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.