Graph Upsampling

Graph upsampling techniques aim to improve the performance of graph neural networks (GNNs) by addressing issues like over-smoothing and information loss during message passing. Current research focuses on developing adaptive and universal upsampling methods, often involving modifications to the graph structure (e.g., adding nodes or edges) or leveraging transformer architectures for more nuanced feature refinement. These advancements are significant because they enhance the accuracy and robustness of GNNs across various applications, including node classification and 3D mesh generation, particularly in scenarios with imbalanced data or complex graph structures. The resulting improvements contribute to more effective analysis of graph-structured data in diverse scientific and engineering domains.

Papers

July 28, 2024

UniGAP: A Universal and Adaptive Graph Upsampling Approach to Mitigate Over-Smoothing in Node Classification Tasks
Xiaotang Wang, Yun Zhu, Haizhou Shi, Yongchao Liu, Chuntao Hong
Graph Transformer Node Classification Universal Image Smoothing Problem Graph Upsampling

March 20, 2024

T-Pixel2Mesh: Combining Global and Local Transformer for 3D Mesh Generation from a Single Image
Shijie Zhang, Boyan Jiang, Keke He, Junwei Zhu, Ying Tai, Chengjie Wang, Yinda Zhang, Yanwei Fu
Fine Grained Single Image World Event Local Transformer Single View 3D Reconstruction Global Shape Mesh Deformation 3D Mesh Generation Graph Upsampling

August 17, 2023

Half-Hop: A graph upsampling approach for slowing down message passing
Mehdi Azabou, Venkataramana Ganesh, Shantanu Thakoor, Chi-Heng Lin, Lakshmi Sathidevi, Ran Liu, Michal Valko, Petar Veličković, Eva L. Dyer
Self Supervised Learning Self Supervised Graph Drawing Graph Structured Data System Message Basin Hopping Graph Upsampling Slow Node

February 5, 2022

Graph Neural Network with Curriculum Learning for Imbalanced Node Classification
Xiaohe Li, Lijie Wen, Yawen Deng, Fuli Feng, Xuming Hu, Lei Wang, Zide Fan
Graph Neural Network Node Classification Curriculum Learning Graph Classification Graph Learning Task Imbalanced Node Classification Graph Upsampling

Graph Upsampling

Papers

UniGAP: A Universal and Adaptive Graph Upsampling Approach to Mitigate Over-Smoothing in Node Classification Tasks

T-Pixel2Mesh: Combining Global and Local Transformer for 3D Mesh Generation from a Single Image

Half-Hop: A graph upsampling approach for slowing down message passing

Graph Neural Network with Curriculum Learning for Imbalanced Node Classification