GNN Layer

A GNN layer is a fundamental building block in Graph Neural Networks (GNNs), designed to process information from graph-structured data by aggregating and transforming node features based on their neighborhood connections. Current research focuses on addressing limitations such as over-smoothing (where deeper networks lose node distinctiveness) and improving efficiency through optimized initialization methods, novel aggregation strategies (like "Aggregation before Communication"), and architectural innovations like incorporating persistent homology or multi-rate gradient gating. These advancements aim to enhance GNN performance across various tasks (node classification, link prediction, graph classification) and enable the training of deeper, more powerful models for applications ranging from healthcare to social network analysis.

Papers

November 5, 2024

DA-MoE: Addressing Depth-Sensitivity in Graph-Level Analysis through Mixture of Experts
Zelin Yao, Chuang Liu, Xianke Meng, Yibing Zhan, Jia Wu, Shirui Pan, Wenbin Hu
Graph Neural Network Mixture Component Expert Knowledge Graph Data Deeper Network GNN Layer Graph Level

May 8, 2024

Conditional Local Feature Encoding for Graph Neural Networks
Yongze Wang, Haimin Zhang, Qiang Wu, Min Xu
Graph Neural Network GNN Model Neural Network Baseline GNN Layer

February 26, 2024

Boosting Graph Pooling with Persistent Homology
Chaolong Ying, Xinjian Zhao, Tianshu Yu
Persistent Homology Graph Pooling GNN Layer Graph Pooling Method Topological Invariance

December 5, 2023

On the Initialization of Graph Neural Networks
Jiahang Li, Yakun Song, Xiang Song, David Paul Wipf
Graph Neural Network New Initialization GNN Layer

November 28, 2023

Syntactic Fusion: Enhancing Aspect-Level Sentiment Analysis Through Multi-Tree Graph Integration
Jane Sunny, Tom Padraig, Roggie Terry, Woods Ali
Dependency Tree Aspect Level Sentiment Analysis GNN Layer Graph Integration

October 14, 2023

Efficient Link Prediction via GNN Layers Induced by Negative Sampling
Yuxin Wang, Xiannian Hu, Quan Gan, Xuanjing Huang, Xipeng Qiu, David Wipf
Link Prediction Node Embeddings Negative Sampling Robust Graph GNN Layer

September 12, 2023

September 4, 2023

Layer-wise training for self-supervised learning on graphs
Oscar Pina, Verónica Vilaplana
Graph Neural Network Self Supervised Learning Graph Drawing Layer Wise GNN Layer Inductive Graph

August 23, 2023

Cached Operator Reordering: A Unified View for Fast GNN Training
Julia Bazinska, Andrei Ivanov, Tal Ben-Nun, Nikoli Dryden, Maciej Besta, Siyuan Shen, Torsten Hoefler
Graph Neural Network Graph Convolutional Network Unified View GNN Training GNN Layer GNN Computation Large Scale GNN

December 11, 2022

ABC: Aggregation before Communication, a Communication Reduction Framework for Distributed Graph Neural Network Training and Effective Partition
Junwei Su
Timely Communication Graph Neural Data Aggregation Large Graph GNN Training Communication Cost GNN Layer GNN Training Framework

October 2, 2022

Gradient Gating for Deep Multi-Rate Learning on Graphs
T. Konstantin Rusch, Benjamin P. Chamberlain, Michael W. Mahoney, Michael M. Bronstein, Siddhartha Mishra
Graph Neural Network Graph Drawing Gating Mechanism GNN Layer

August 18, 2022

GraTO: Graph Neural Network Framework Tackling Over-smoothing with Neural Architecture Search
Xinshun Feng, Herun Wan, Shangbin Feng, Hongrui Wang, Jun Zhou, Qinghua Zheng, Minnan Luo
Graph Neural Network Neural Architecture Search Undisciplined Over Smoothing Smoothing Problem GNN Layer Smooth Representation

June 11, 2022

Soft-mask: Adaptive Substructure Extractions for Graph Neural Networks
Mingqi Yang, Yanming Shen, Heng Qi, Baocai Yin
Graph Neural Network Graph Representation Hierarchical Representation Graph Level Task GNN Layer Soft Mask Structure Information Extraction Task Specific Structure

February 4, 2022

Graph-Coupled Oscillator Networks
T. Konstantin Rusch, Benjamin P. Chamberlain, James Rowbottom, Siddhartha Mishra, Michael M. Bronstein
Multi Layer Neural Network GNN Layer