Deep Graph Model

Deep graph models, primarily graph neural networks (GNNs) and graph transformers, aim to leverage the power of graph-structured data by learning representations from interconnected nodes and edges. Current research focuses on addressing challenges like the difficulty of training very deep GNNs, improving their performance in few-shot learning scenarios with limited labeled data, and developing more efficient and robust models, including those that effectively handle diverse node feature types. These advancements are crucial for various applications, such as drug discovery, where deep graph models are used to predict interactions in biological networks, and generally improving the efficiency and interpretability of graph-based machine learning.

Papers

August 7, 2024

Beyond Over-smoothing: Uncovering the Trainability Challenges in Deep Graph Neural Networks
Jie Peng, Runlin Lei, Zhewei Wei
Graph Neural Network Undisciplined Over Smoothing Deep Graph Model

July 10, 2024

A deep graph model for the signed interaction prediction in biological network
Shuyi Jin, Mengji Zhang, Meijie Wang, Lun Yu
Interaction Prediction Biological Network Drug Repurposing Deep Graph Model Biomedical Network

February 3, 2024

Neural Scaling Laws on Graphs
Jingzhe Liu, Haitao Mao, Zhikai Chen, Tong Zhao, Neil Shah, Jiliang Tang
Graph Drawing Graph Transformer Large Graph Neural Scaling Law Model Scaling Data Scaling Deep Graph Model

September 30, 2022

Contrastive Graph Few-Shot Learning
Chunhui Zhang, Hongfu Liu, Jundong Li, Yanfang Ye, Chuxu Zhang
LeArning Abstract Contrastive Learning Shot Learning Contrastive Graph Contrastive Distillation Deep Graph Model Shot Graph

June 16, 2022

A Robust Stacking Framework for Training Deep Graph Models with Multifaceted Node Features
Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Tom Goldstein, David Wipf
Graph Neural Network Layer Wise Graph Datasets Node Feature Graph Propagation Stacking Framework Deep Graph Model

May 24, 2022