ABSTRACT
Real-world graphs exhibit diverse structures, including homophilic and heterophilic patterns, necessitating the development of a universal Graph Contrastive Learning (GCL) framework. Nonetheless, the existing GCLs, especially those with a local focus, lack universality due to the mismatch between the input graph structure and the homophily assumption for two primary components of GCLs. Firstly, the encoder, commonly Graph Convolution Network (GCN), operates as a low-pass filter, which assumes the input graph to be homophilic. This makes it challenging to aggregate features from neighbor nodes of the same class on heterophilic graphs. Secondly, the local positive sampling regards neighbor nodes as positive samples, which is inspired by the homophily assumption. This results in feature similarity amplification for the samples from the different classes (i.e., FALSE positive samples). Therefore, it is crucial to feed the encoder and positive sampling of GCLs with homophilic graph structures. This paper presents a novel GCL framework, named gRaph cOntraStive Exploring uNiversality (ROSEN), designed to achieve this objective. Specifically, ROSEN equips a local graph structure inference module, utilizing the Block Diagonal Property (BDP) of the affinity matrix extracted from node ego networks. This module can generate the homophilic graph structure by selectively removing disassortative edges. Extensive evaluations validate the effectiveness and universality of ROSEN across node classification and node clustering tasks.
Supplemental Material
- Kristen M Altenburger and Johan Ugander. 2018. Monophily in social networks introduces similarity among friends-of-friends. Nature human behaviour, Vol. 2, 4 (2018), 284--290.Google Scholar
- Stephen P. Boyd and Lieven Vandenberghe. 2014. Convex Optimization.Google Scholar
- Jingfan Chen, Guanghui Zhu, Yifan Qi, Chunfeng Yuan, and Yihua Huang. 2022. Towards Self-supervised Learning on Graphs with Heterophily. In CIKM. ACM, 201--211.Google Scholar
- Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey E. Hinton. 2020. A Simple Framework for Contrastive Learning of Visual Representations. In ICML. 1597--1607.Google Scholar
- Fan RK Chung and Fan Chung Graham. 1997. Spectral graph theory. Number 92.Google Scholar
- Aditya Grover and Jure Leskovec. 2016. node2vec: Scalable Feature Learning for Networks. In SIGKDD. 855--864.Google Scholar
- John A Hartigan and Manchek A Wong. 1979. Algorithm AS 136: A k-means clustering algorithm. Journal of the royal statistical society. series c (applied statistics), Vol. 28, 1 (1979), 100--108.Google ScholarCross Ref
- Kaveh Hassani and Amir Hosein Khas Ahmadi. 2020. Contrastive Multi-View Representation Learning on Graphs. In ICML. 4116--4126.Google Scholar
- Dongxiao He, Chundong Liang, Huixin Liu, Mingxiang Wen, Pengfei Jiao, and Zhiyong Feng. 2022. Block Modeling-Guided Graph Convolutional Neural Networks. In AAAI. 4022--4029.Google Scholar
- Dongxiao He, Jitao Zhao, Rui Guo, Zhiyong Feng, Di Jin, Yuxiao Huang, Zhen Wang, and Weixiong Zhang. 2023. Contrastive Learning Meets Homophily: Two Birds with One Stone. In ICML. 12775--12789.Google Scholar
- Roger A Horn and Charles R Johnson. 2012. Matrix analysis. Cambridge university press.Google ScholarDigital Library
- Zhenyu Hou, Xiao Liu, Yukuo Cen, Yuxiao Dong, Hongxia Yang, Chunjie Wang, and Jie Tang. 2022. GraphMAE: Self-Supervised Masked Graph Autoencoders. In KDD. 594--604.Google Scholar
- Weihua Hu, Matthias Fey, Marinka Zitnik, Yuxiao Dong, Hongyu Ren, Bowen Liu, Michele Catasta, and Jure Leskovec. 2020. Open Graph Benchmark: Datasets for Machine Learning on Graphs. In NeurIPS.Google Scholar
- Diederik P. Kingma and Jimmy Ba. 2015. Adam: A Method for Stochastic Optimization. In ICLR.Google Scholar
- Thomas N. Kipf and Max Welling. 2016. Variational Graph Auto-Encoders. CoRR (2016).Google Scholar
- Thomas N. Kipf and Max Welling. 2017. Semi-Supervised Classification with Graph Convolutional Networks. In ICLR.Google Scholar
- Wen-Zhi Li, Chang-Dong Wang, Hui Xiong, and Jian-Huang Lai. 2023. HomoGCL: Rethinking Homophily in Graph Contrastive Learning. In SIGKDD. 1341--1352.Google Scholar
- Canyi Lu, Jiashi Feng, Zhouchen Lin, Tao Mei, and Shuicheng Yan. 2019. Subspace Clustering by Block Diagonal Representation. IEEE Trans. Pattern Anal. Mach. Intell., Vol. 41, 2 (2019), 487--501.Google ScholarDigital Library
- Yao Ma, Xiaorui Liu, Tong Zhao, Yozen Liu, Jiliang Tang, and Neil Shah. 2020. A Unified View on Graph Neural Networks as Graph Signal Denoising. arxiv: 2010.01777Google Scholar
- Miller McPherson, Lynn Smith-Lovin, and James M Cook. 2001. Birds of a feather: Homophily in social networks. Annual review of sociology, Vol. 27, 1 (2001), 415--444.Google Scholar
- Pé ter Mernyei and Catalina Cangea. 2020. Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networks. CoRR (2020).Google Scholar
- Hongbin Pei, Bingzhe Wei, Kevin Chen-Chuan Chang, Yu Lei, and Bo Yang. 2020. Geom-GCN: Geometric Graph Convolutional Networks. In ICLR.Google Scholar
- Bryan Perozzi, Rami Al-Rfou, and Steven Skiena. 2014. DeepWalk: online learning of social representations. In SIGKDD. 701--710.Google Scholar
- Oleg Platonov, Denis Kuznedelev, Michael Diskin, Artem Babenko, and Liudmila Prokhorenkova. 2023. A critical look at the evaluation of GNNs under heterophily: Are we really making progress?. In ICLR.Google Scholar
- Benedek Rozemberczki, Carl Allen, and Rik Sarkar. 2021. Multi-Scale attributed node embedding. J. Complex Networks (2021).Google Scholar
- Prithviraj Sen, Galileo Namata, Mustafa Bilgic, Lise Getoor, Brian Gallagher, and Tina Eliassi-Rad. 2008. Collective Classification in Network Data. AI Mag. (2008), 93--106.Google Scholar
- Oleksandr Shchur, Maximilian Mumme, Aleksandar Bojchevski, and Stephan Gü nnemann. 2018. Pitfalls of Graph Neural Network Evaluation. CoRR, Vol. abs/1811.05868 (2018).Google Scholar
- Jie Tang, Jimeng Sun, Chi Wang, and Zi Yang. 2009. Social influence analysis in large-scale networks. In KDD. ACM, 807--816.Google Scholar
- Shantanu Thakoor, Corentin Tallec, Mohammad Gheshlaghi Azar, Rémi Munos, Petar Velivc ković, and Michal Valko. 2021. Bootstrapped representation learning on graphs. (2021).Google Scholar
- A"a ron van den Oord, Yazhe Li, and Oriol Vinyals. 2018. Representation Learning with Contrastive Predictive Coding. CoRR, Vol. abs/1807.03748 (2018).Google Scholar
- Petar Velickovic, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. 2018. Graph Attention Networks. In ICLR.Google Scholar
- Petar Velickovic, William Fedus, William L. Hamilton, Pietro Liò, Yoshua Bengio, and R. Devon Hjelm. 2019. Deep Graph Infomax. In ICLR.Google Scholar
- Haonan Wang, Jieyu Zhang, Qi Zhu, and Wei Huang. 2022. Can Single-Pass Contrastive Learning Work for Both Homophilic and Heterophilic Graph? CoRR, Vol. abs/2211.10890.Google Scholar
- Yu Wang, Wotao Yin, and Jinshan Zeng. 2019. Global convergence of ADMM in nonconvex nonsmooth optimization. Journal of Scientific Computing, Vol. 78 (2019), 29--63.Google ScholarDigital Library
- Yifei Wang, Qi Zhang, Tianqi Du, Jiansheng Yang, Zhouchen Lin, and Yisen Wang. 2023. A Message Passing Perspective on Learning Dynamics of Contrastive Learning. In ICLR. OpenReview.net.Google Scholar
- Yaochen Xie, Zhao Xu, Jingtun Zhang, Zhengyang Wang, and Shuiwang Ji. 2022. Self-supervised learning of graph neural networks: A unified review. IEEE transactions on pattern analysis and machine intelligence, Vol. 45, 2 (2022), 2412--2429.Google Scholar
- Keyulu Xu, Chengtao Li, Yonglong Tian, Tomohiro Sonobe, Ken-ichi Kawarabayashi, and Stefanie Jegelka. 2018. Representation Learning on Graphs with Jumping Knowledge Networks. In ICML. 5449--5458.Google Scholar
- Yujun Yan, Milad Hashemi, Kevin Swersky, Yaoqing Yang, and Danai Koutra. 2021. Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks. arxiv: 2102.06462 [cs.LG]Google Scholar
- Liang Yang, Cheng Chen, Weixun Li, Bingxin Niu, Junhua Gu, Chuan Wang, Dongxiao He, Yuanfang Guo, and Xiaochun Cao. 2022. Self-Supervised Graph Neural Networks via Diverse and Interactive Message Passing. In AAAI. 4327--4336.Google Scholar
- Hengrui Zhang, Qitian Wu, Yu Wang, Shaofeng Zhang, Junchi Yan, and Philip S. Yu. 2022. Localized Contrastive Learning on Graphs. CoRR (2022).Google Scholar
- Jiong Zhu, Ryan A. Rossi, Anup Rao, Tung Mai, Nedim Lipka, Nesreen K. Ahmed, and Danai Koutra. 2021a. Graph Neural Networks with Heterophily. In AAAI. 11168--11176.Google Scholar
- Yanqiao Zhu, Yichen Xu, Feng Yu, Qiang Liu, Shu Wu, and Liang Wang. 2020. Deep Graph Contrastive Representation Learning. CoRR, Vol. abs/2006.04131 (2020). arxiv: 2006.04131Google Scholar
- Yanqiao Zhu, Yichen Xu, Feng Yu, Qiang Liu, Shu Wu, and Liang Wang. 2021b. Graph Contrastive Learning with Adaptive Augmentation. In WWW.Google Scholar
Index Terms
- Graph Contrastive Learning Reimagined: Exploring Universality
Recommendations
Cross-view graph contrastive learning with hypergraph
AbstractGraph contrastive learning (GCL) provides a new perspective to alleviate the reliance on labeled data for graph representation learning. Recent efforts on GCL leverage various graph augmentation strategies, i.e., node dropping and edge masking, ...
Highlights- We proposed that hypergraphs are used as a paradigm to enhance graph contrastive learning.
- We propose a novel diffusion model-based fusion mechanism that aligns the positive examples.
- Our experimental results all exceed existing ...
Label-guided graph contrastive learning for semi-supervised node classification
AbstractSemi-supervised node classification is a task of predicting the labels of unlabeled nodes using limited labeled nodes and numerous unlabeled nodes. Recently, Graph Neural Networks (GNNs) have achieved remarkable success in this task. However, ...
Highlights- The framework explores semantic-level feature similarity.
- The self-checking mechanism ensures the authenticity of the positive nodes.
- The reweighting strategy enhances the effect of hard negative nodes.
- The training algorithm ...
Learning to Augment Graph Structure for both Homophily and Heterophily Graphs
Machine Learning and Knowledge Discovery in Databases: Research TrackAbstractRecent years have witnessed great successes in performing graph structure learning for Graph Neural Networks (GNNs). However, comparatively little work studies structure augmentation for graphs, where the augmented structures are only used for ...
Comments