Multi-Augmentation Contrastive Learning as Multi-Objective Optimization for Graph Neural Networks

被引:0
|
作者
Li, Xu [1 ]
Chen, Yongsheng [1 ]
机构
[1] Tongji Univ, Shanghai, Peoples R China
关键词
graph neural networks; multi-objective Learning; self-supervised learning;
D O I
10.1007/978-3-031-33377-4_38
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recently self-supervised learning is gaining popularity for Graph Neural Networks (GNN) by leveraging unlabeled data. Augmentation plays a key role in self-supervision. While there is a common set of image augmentation methods that preserve image labels in general, graph augmentation methods do not guarantee consistent graph semantics and are usually domain dependent. Existing self-supervised GNN models often handpick a small set of augmentation techniques that limit the performance of the model. In this paper, we propose a common set of graph augmentation methods to a wide range of GNN tasks, and rely on the Pareto optimality to select and balance among these possibly conflicting augmented versions, called Pareto Graph Contrastive Learning (PGCL) framework. We show that while random selection of the same set of augmentation leads to slow convergence or even divergence, PGCL converges much faster with lower error rate. Extensive experiments on multiple datasets of different domains and scales demonstrate superior or comparable performance of PGCL.
引用
收藏
页码:495 / 507
页数:13
相关论文
共 50 条
  • [41] Convolutional neural networks optimization using multi-objective particle swarm optimization algorithm
    Rashno, Armin
    Fadaei, Sadegh
    INFORMATION SCIENCES, 2025, 689
  • [42] Graph Learning Assisted Multi-Objective Integer Programming
    Wu, Yaoxin
    Song, Wen
    Cao, Zhiguang
    Zhang, Jie
    Gupta, Abhishek
    Lin, Mingyan Simon
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [43] Federated Learning Meets Multi-Objective Optimization
    Hu, Zeou
    Shaloudegi, Kiarash
    Zhang, Guojun
    Yu, Yaoliang
    IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING, 2022, 9 (04): : 2039 - 2051
  • [44] Multi-objective optimization of a dimpled channel for heat transfer augmentation
    Samad, Abdus
    Lee, Ki-Don
    Kim, Kwang-Yong
    HEAT AND MASS TRANSFER, 2008, 45 (02) : 207 - 217
  • [45] Multi-objective optimization of a dimpled channel for heat transfer augmentation
    Abdus Samad
    Ki-Don Lee
    Kwang-Yong Kim
    Heat and Mass Transfer, 2008, 45
  • [46] Modeling and multi-objective optimization of square cyclones using CFD and neural networks
    Safikhani, H.
    Akhavan-Behabadi, M. A.
    Nariman-Zadeh, N.
    Abadi, M. J. Mahmood
    CHEMICAL ENGINEERING RESEARCH & DESIGN, 2011, 89 (3A): : 301 - 309
  • [47] Multi-objective optimization for modular granular neural networks applied to pattern recognition
    Melin, Patricia
    Sanchez, Daniela
    INFORMATION SCIENCES, 2018, 460 : 594 - 610
  • [48] Evolving ARTMAP Neural Networks Using Multi-Objective Particle Swarm Optimization
    Granger, Eric
    Prieur, Donavan
    Connolly, Jean-Francois
    2010 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2010,
  • [49] DeepMaker: A multi-objective optimization framework for deep neural networks in embedded systems
    Loni, Mohammad
    Sinaei, Sima
    Zoljodi, Ali
    Daneshtalab, Masoud
    Sjodin, Mikael
    MICROPROCESSORS AND MICROSYSTEMS, 2020, 73 (73)
  • [50] Application of Multi-Objective optimization algorithm and Artificial Neural Networks at machining process
    Jafarian, Farshid
    Amirabadi, Hossein
    Sadri, Javad
    2013 FIRST IRANIAN CONFERENCE ON PATTERN RECOGNITION AND IMAGE ANALYSIS (PRIA), 2013,