图，如社会网络和分子图，是现实世界中无处不在的数据结构。由于它们的普遍存在，从图结构数据中提取有意义的模式以方便下游任务的开展具有重要的研究意义。图表示学习取代了手工设计的特征，它可以学习能够编码关于图的丰富信息的表示。它在节点分类、链路预测、图分类等任务中取得了巨大的成功，近年来受到越来越多的关注。近年来，由于图结构的强大表现力，用机器学习方法分析图的研究越来越受到重视。图神经网络（GNN）是一类基于深度学习的处理图域信息的方法。2021年年初，图神经网络一举成为热门研究主题。AAAI2021年,录用了165篇论文。
到目前，相关研究的已经非常多了，不过我们回过头来看思考和回顾一下：解决图(Graph)数据的方法和思路有那些，这些方法和思路跟图像文本有什么异同？各大顶会提出的图网络模型有什么样的基本假设和主要功能？本文整理了15篇图神经网络(GNN)相关的文献综述，这15篇相关的文献综述从不同的角度对图神经网络的发展进行了详细的总结和概述。下面是本文整理的列表:
1. Deep Learning on Graphs: A Survey
清华大学 Ziwei Zhang, Peng Cui, Wenwu Zhu
2. Graph Neural Networks: A Review of Methods and Applications
清华大学Jie Zhou, Ganqu Cui, Zhengyan Zhang, Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, Maosong Sun
3. A Comprehensive Survey on Graph Neural Networks
UTS等 Zonghan Wu, Shirui Pan, Fengwen Chen, Guodong Long, Chengqi Zhang, Philip S. Yu
4. A Gentle Introduction to Deep Learning for Graphs
University of Pisa Davide Bacciua , Federico Erricaa , Alessio Michelia , Marco Poddaa
5. LEARNING REPRESENTATIONS OF GRAPH DATA: A SURVEY
UCL Mital Kinderkhedia
6. Graph Neural Networks for Small Graph and Giant Network Representation Learning: An Overview
Information Fusion and Mining Laboratory Jiawei Zhang
7. A Survey on The Expressive Power of Graph Neural Networks
东京大学 Ryoma Sato
8. Relational inductive biases, deep learning, and graph networks
DeepMind;等
9. Adversarial Attack and Defense on Graph Data: A Survey
UIUC等Lichao Sun, Yingtong Dou, Carl Yang等
10. Heterogeneous Network Representation Learning: Survey, Benchmark, Evaluation, and Beyond
UIUC Carl Yang, Yuxin Xiao, Yu Zhang, Yizhou Sun, and Jiawei Han,
11 Automated Machine Learning on Graphs: A Survey
清华大学 Ziwei Zhang, Xin Wang, Wenwu Zhu
12. Graph Self-Supervised Learning: A Survey
Monash University等 Yixin Liu1 , Shirui Pan1 , Ming Jin1 , Chuan Zhou2 , Feng Xia3 , Philip S. Yu
13. Self-Supervised Learning of Graph Neural Networks: A Unified Review
Texas A&M University等 Yaochen Xie, Zhao Xu, Zhengyang Wang, Shuiwang Ji
14. Meta-Learning with Graph Neural Networks: Methods and Applications
Columbia University等
Debmalya Mandal, Sourav Medya, Brian Uzzi, Charu Aggarwal
15. Deep Graph Structure Learning for Robust Representations: A Survey
中科院自动化研究所：Yanqiao Zhu∗ , Weizhi Xu∗ , Jinghao Zhang∗ , Qiang Liu , Shu Wu† and Liang Wang
