Rethinking the Graph Polynomial Filter via Positive and Negative Coupling Analysis

arXiv:2404.10353v1 Announce Type: new
Abstract: Recently, the optimization of polynomial filters within Spectral Graph Neural Networks (GNNs) has emerged as a prominent research focus. Existing spectral GNNs mainly emphasize polynomial properties in filter design, introducing computational overhead and neglecting the integration of crucial graph structure information. We argue that incorporating graph information into basis construction can enhance understanding of polynomial basis, and further facilitate simplified polynomial filter design. Motivated by this, we first propose a Positive and Negative Coupling Analysis …

abstract analysis arxiv computational cs.lg cs.si design filter filters focus gnns graph graph neural networks information integration negative networks neural networks optimization polynomial positive research the graph type via