Spectral GNN

Spectral Graph Neural Networks (GNNs) leverage the spectral properties of graphs to learn representations for nodes and graphs, aiming to improve efficiency and performance compared to spatial GNNs. Current research focuses on developing novel spectral filters and Laplacian matrices to handle diverse graph types (directed, signed, weighted) and address limitations like over-smoothing and information loss, leading to architectures such as ChebNet, Specformer, and models employing learnable orthonormal bases or framelets. These advancements enhance the applicability of GNNs to various domains, including medical image segmentation, computational mechanics, and human pose estimation, by improving accuracy and efficiency in tasks such as node classification, link prediction, and graph-level tasks.

Papers

October 15, 2024

Point-Calibrated Spectral Neural Operators
Xihang Yue, Linchao Zhu, Yi Yang
Spectral Learning Spectral GNN Spectral Neural Network Spectral Space

September 13, 2024

Spectral U-Net: Enhancing Medical Image Segmentation via Spectral Decomposition
Yaopeng Peng, Milan Sonka, Danny Z. Chen
Medical Image Segmentation Liver Tumor Segmentation Spectral GNN Spectral Decomposition Complex Wavelet Free Down Sampling Operation

September 1, 2024

Spatio-spectral graph neural operator for solving computational mechanics problems on irregular domain and unstructured grid
Subhankar Sarkar, Souvik Chakraborty
Neural Operator Operator Learning Solid Mechanic Spectral Graph Neural Network Spectral Graph Spectral GNN Unstructured Grid Spatial Graph Irregular Domain

June 3, 2024

Equivariant Machine Learning on Graphs with Nonlinear Spectral Filters
Ya-Wei Eileen Lin, Ronen Talmon, Ron Levie
Convolutional Neural Network Graph Drawing Spectral GNN Equivariant Machine Learning Graph Shift Operator Shift Equivariance Spectral Magnitude

April 6, 2024

Spectral GNN via Two-dimensional (2-D) Graph Convolution
Guoming Li, Jian Yang, Shangsong Liang, Dongsheng Luo
2 Dimensional Graph Convolution Spectral Graph Neural Network Spectral GNN Spectral Graph Convolution

December 28, 2023

Graph Learning in 4D: a Quaternion-valued Laplacian to Enhance Spectral GCNs
Stefano Fiorini, Stefano Coniglio, Michele Ciavotta, Enza Messina
Graph Learning 4 Dimensional Spectral GNN Spectral Graph Convolution Quaternion Valued Laplacian Matrix Quaternion Matrix

October 8, 2023

Simplifying GNN Performance with Low Rank Kernel Models
Luciano Vinas, Arash A. Amini
GNN Based Semi Supervised Node Classification Spectral GNN GNN Performance Low Rank Kernel

March 24, 2023

LON-GNN: Spectral GNNs with Learnable Orthonormal Basis
Qian Tao, Zhen Wang, Wenyuan Yu, Yaliang Li, Zhewei Wei
Spectral Graph Neural Network Spectral Graph Spectral GNN Polynomial Basis

March 2, 2023

Specformer: Spectral Graph Neural Networks Meet Transformers
Deyu Bo, Chuan Shi, Lele Wang, Renjie Liao
Spectral Graph Neural Network Spectral GNN Spectral Graph Convolution

February 24, 2023

Graph Neural Networks with Learnable and Optimal Polynomial Bases
Yuhe Guo, Zhewei Wei
Graph Neural Network Spectral GNN Polynomial Filter Polynomial Basis

October 29, 2022

Clenshaw Graph Neural Networks
Yuhe Guo, Zhewei Wei
Graph Representation Spectral GNN Graph Spectrum

October 20, 2022

A Magnetic Framelet-Based Convolutional Neural Network for Directed Graphs
Lequan Lin, Junbin Gao
Directed Graph Spectral GNN Framelet Transform Spectral Graph Convolution Graph Framelets

September 1, 2022

MSGNN: A Spectral Graph Neural Network Based on a Novel Magnetic Signed Laplacian
Yixuan He, Michael Permultter, Gesine Reinert, Mihai Cucuringu
Spectral Graph Neural Network Spectral Graph Spectral GNN Synthetic Network Permanent Magnet Laplacian Canonization

May 27, 2022

EvenNet: Ignoring Odd-Hop Neighbors Improves Robustness of Graph Neural Networks
Runlin Lei, Zhen Wang, Yaliang Li, Bolin Ding, Zhewei Wei
Graph Neural Network Native Robustness Graph Machine Learning Spectral GNN Hop Neighbor Robust SNN Polynomial Graph Filter

May 26, 2022

SigMaNet: One Laplacian to Rule Them All
Stefano Fiorini, Stefano Coniglio, Michele Ciavotta, Enza Messina
Spectral GNN Laplacian Canonization

May 23, 2022

How Powerful are Spectral Graph Neural Networks
Xiyuan Wang, Muhan Zhang
Graph Neural Network Graph Signal Spectral Graph Neural Network Spectral GNN

April 12, 2022

Adaptive Cross-Attention-Driven Spatial-Spectral Graph Convolutional Network for Hyperspectral Image Classification
Jin-Yu Yang, Heng-Chao Li, Wen-Shuai Hu, Lei Pan, Qian Du
Deep Convolutional Neural Network Adaptive Importance Hyperspectral Image Classification Spectral GNN Spectral Attention Spectral Graph Convolution

March 20, 2022

3D Human Pose Estimation Using M\"obius Graph Convolutional Networks
Niloofar Azizi, Horst Possegger, Emanuele Rodolà, Horst Bischof
Convolutional Neural Network 3D Human 3D Human Pose Estimation Spectral GNN Pose Estimation Benchmark M\"obius Transform

February 16, 2022

When Does A Spectral Graph Neural Network Fail in Node Classification?
Zhixian Chen, Tengfei Ma, Yang Wang
Node Classification Graph Learning Task Graph Filter Spectral Graph Neural Network Spectral GNN

February 4, 2022

Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited
Mingguo He, Zhewei Wei, Ji-Rong Wen
Convolutional Neural Network Graph Drawing Graph Convolution Spectral GNN Spectral Graph Convolution Chebyshev Davidson Algorithm