Graph Neural Network
Graph Neural Networks (GNNs) are a class of machine learning models designed to analyze and learn from data represented as graphs, focusing on capturing relationships between nodes and their impact on downstream tasks like node classification and link prediction. Current research emphasizes improving GNN performance by addressing limitations such as oversmoothing and oversquashing through architectural innovations (e.g., incorporating residual connections, Cayley graph propagation) and novel training techniques (e.g., contrastive learning, Laplacian regularization). GNNs are proving valuable across diverse fields, including social network analysis, drug discovery, and financial modeling, offering powerful tools for analyzing complex relational data where traditional methods fall short.
Papers
Imbalanced Graph-Level Anomaly Detection via Counterfactual Augmentation and Feature Learning
Zitong Wang, Xuexiong Luo, Enfeng Song, Qiuqing Bai, Fu Lin
Biased Backpressure Routing Using Link Features and Graph Neural Networks
Zhongyuan Zhao, Bojan Radojičić, Gunjan Verma, Ananthram Swami, Santiago Segarra
The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challenges
Sitao Luan, Chenqing Hua, Qincheng Lu, Liheng Ma, Lirong Wu, Xinyu Wang, Minkai Xu, Xiao-Wen Chang, Doina Precup, Rex Ying, Stan Z. Li, Jian Tang, Guy Wolf, Stefanie Jegelka
The $μ\mathcal{G}$ Language for Programming Graph Neural Networks
Matteo Belenchia, Flavio Corradini, Michela Quadrini, Michele Loreti
The Effectiveness of Curvature-Based Rewiring and the Role of Hyperparameters in GNNs Revisited
Floriano Tori, Vincent Holst, Vincent Ginis
Graph Neural Network Causal Explanation via Neural Causal Models
Arman Behnam, Binghui Wang
TinyGraph: Joint Feature and Node Condensation for Graph Neural Networks
Yezi Liu, Yanning Shen
Explaining Graph Neural Networks for Node Similarity on Graphs
Daniel Daza, Cuong Xuan Chu, Trung-Kien Tran, Daria Stepanova, Michael Cochez, Paul Groth
STAGE: Simplified Text-Attributed Graph Embeddings Using Pre-trained LLMs
Aaron Zolnai-Lucas, Jack Boylan, Chris Hokamp, Parsa Ghaffari
Graph Neural Networks and Spatial Information Learning for Post-Processing Ensemble Weather Forecasts
Moritz Feik, Sebastian Lerch, Jan Stühmer
Thermodynamics-Consistent Graph Neural Networks
Jan G. Rittig, Alexander Mitsos
Greener GRASS: Enhancing GNNs with Encoding, Rewiring, and Attention
Tongzhou Liao, Barnabás Póczos