GiGaMAE: Generalizable Graph Masked Autoencoder via Collaborative Latent Space Reconstruction

被引:10
|
作者
Shi, Yucheng [1 ]
Dong, Yushun [2 ]
Tan, Qiaoyu [3 ]
Li, Jundong [2 ]
Liu, Ninghao [1 ]
机构
[1] Univ Georgia, Athens, GA USA
[2] Univ Virginia, Charlottesville, VA USA
[3] Texas A&M Univ, College Stn, TX USA
关键词
Self-supervised Learning; Graph Mining; Masked Autoencoder;
D O I
10.1145/3583780.3614894
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Self-supervised learning with masked autoencoders has recently gained popularity for its ability to produce effective image or textual representations, which can be applied to various downstream tasks without retraining. However, we observe that the current masked autoencoder models lack good generalization ability on graph data. To tackle this issue, we propose a novel graph masked autoencoder framework called GiGaMAE. Different from existing masked autoencoders that learn node presentations by explicitly reconstructing the original graph components (e.g., features or edges), in this paper, we propose to collaboratively reconstruct informative and integrated latent embeddings. By considering embeddings encompassing graph topology and attribute information as reconstruction targets, our model could capture more generalized and comprehensive knowledge. Furthermore, we introduce a mutual information based reconstruction loss that enables the effective reconstruction of multiple targets. This learning objective allows us to differentiate between the exclusive knowledge learned from a single target and common knowledge shared by multiple targets. We evaluate our method on three downstream tasks with seven datasets as benchmarks. Extensive experiments demonstrate the superiority of GiGaMAE against state-of-the-art baselines. We hope our results will shed light on the design of foundation models on graph-structured data. Our code is available at: https://github.com/sycny/GiGaMAE.
引用
收藏
页码:2259 / 2269
页数:11
相关论文
共 20 条
  • [1] Advancing Latent Representation Ranking for Masked Graph Autoencoder
    Hu, Yulan
    Chen, Ge
    Ouyang, Sheng
    Yang, Zhirui
    Wan, Junchen
    Zhang, Fuzheng
    Wang, Zhongyuan
    Cao, Zhao
    Wu, Shangquan
    Liu, Yong
    DATABASE SYSTEMS FOR ADVANCED APPLICATIONS, PT VI, DASFAA 2024, 2024, 14855 : 385 - 394
  • [2] Critical temperature of the classical XY model via autoencoder latent space sampling
    Willnecker, Brandon
    Moodley, Mervlyn
    PHYSICAL REVIEW E, 2025, 111 (01)
  • [3] Latent space improved masked reconstruction model for human skeleton-based action recognition
    Chen, Enqing
    Wang, Xueting
    Guo, Xin
    Zhu, Ying
    Li, Dong
    Frontiers in Neurorobotics, 2024, 19
  • [4] MaskCAE: Masked Convolutional AutoEncoder via Sensor Data Reconstruction for Self-Supervised Human Activity Recognition
    Cheng, Dongzhou
    Zhang, Lei
    Qin, Lutong
    Wang, Shuoyuan
    Wu, Hao
    Song, Aiguo
    IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2024, 28 (05) : 2687 - 2698
  • [5] Generating and Blending Game Levels via Quality-Diversity in the Latent Space of a Variational Autoencoder
    Sarkar, Anurag
    Cooper, Seth
    PROCEEDINGS OF THE 16TH INTERNATIONAL CONFERENCE ON THE FOUNDATIONS OF DIGITAL GAMES, FDG 2021, 2021,
  • [6] Fault Detection via Autoencoder Latent Space Differences Between Reference Model and the Plant Operation
    Villagomez, Enrique Luna
    Mahyar, Hamidreza
    Mahalec, Vladimir
    IFAC PAPERSONLINE, 2024, 58 (14): : 628 - 633
  • [7] Reconstruction and Generation of Porous Metamaterial Units Via Variational Graph Autoencoder and Large Language Model
    Khanghah, Kiarash Naghavi
    Wang, Zihan
    Xu, Hongyi
    JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING, 2025, 25 (02)
  • [8] Multi-view latent space learning framework via adaptive graph embedding
    Li, Haohao
    Li, Zhong
    Wang, Huibing
    JOURNAL OF ELECTRONIC IMAGING, 2024, 33 (06)
  • [9] Denoising Point Clouds in Latent Space via Graph Convolution and Invertible Neural Network
    Mao, Aihua
    Yan, Biao
    Ma, Zijing
    He, Ying
    2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2024, 2024, : 5768 - 5777
  • [10] Probabilistic Graph Convolutional Network via Topology-Constrained Latent Space Model
    Yang, Liang
    Guo, Yuanfang
    Gu, Junhua
    Jin, Di
    Yang, Bo
    Cao, Xiaochun
    IEEE TRANSACTIONS ON CYBERNETICS, 2022, 52 (04) : 2123 - 2136