Generalized Laplacian Regularized Framelet Graph Neural Networks
This work addresses graph signal processing problems for researchers in machine learning, but it appears incremental as it builds on existing p-Laplacian and framelet methods.
The paper tackles graph learning tasks by introducing two novel models, pL-UFG and pL-fUFG, which combine p-Laplacian with framelet graph convolutions, and reports excellent performance in node classification and signal denoising.
This paper introduces a novel Framelet Graph approach based on p-Laplacian GNN. The proposed two models, named p-Laplacian undecimated framelet graph convolution (pL-UFG) and generalized p-Laplacian undecimated framelet graph convolution (pL-fUFG) inherit the nature of p-Laplacian with the expressive power of multi-resolution decomposition of graph signals. The empirical study highlights the excellent performance of the pL-UFG and pL-fUFG in different graph learning tasks including node classification and signal denoising.