Embedding Compression with Hashing for Efficient Representation Learning in Large-Scale Graph

被引:7
|
作者
Yeh, Chin-Chia Michael [1 ]
Gu, Mengting [1 ]
Zheng, Yan [1 ]
Chen, Huiyuan [1 ]
Ebrahimi, Javid [1 ]
Zhuang, Zhongfang [1 ]
Wang, Junpeng [1 ]
Wang, Liang [1 ]
Zhang, Wei [1 ]
机构
[1] Visa Res, Austin, TX 78759 USA
关键词
graph neural network; compression; low-bit embeddings;
D O I
10.1145/3534678.3539068
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Graph neural networks (GNNs) are deep learning models designed specifically for graph data, and they typically rely on node features as the input to the first layer. When applying such a type of network on the graph without node features, one can extract simple graph-based node features (e.g., number of degrees) or learn the input node representations (i.e., embeddings) when training the network. While the latter approach, which trains node embeddings, more likely leads to better performance, the number of parameters associated with the embeddings grows linearly with the number of nodes. It is therefore impractical to train the input node embeddings together with GNNs within graphics processing unit (GPU) memory in an end-to-end fashion when dealing with industrial-scale graph data. Inspired by the embedding compression methods developed for natural language processing (NLP) tasks, we develop a node embedding compression method where each node is compactly represented with a bit vector instead of a floating-point vector. The parameters utilized in the compression method can be trained together with GNNs. We show that the proposed node embedding compression method achieves superior performance compared to the alternatives.
引用
收藏
页码:4391 / 4401
页数:11
相关论文
共 50 条
  • [1] Efficient Supervised Graph Embedding Hashing for large-scale cross-media retrieval
    Yao, Tao
    Wang, Ruxin
    Wang, Jintao
    Li, Ying
    Yue, Jun
    Yan, Lianshan
    Tian, Qi
    [J]. PATTERN RECOGNITION, 2024, 145
  • [2] Large-scale Graph Representation Learning
    Leskovec, Jure
    [J]. 2017 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2017, : 4 - 4
  • [3] Large-scale knowledge graph representation learning
    Badrouni, Marwa
    Katar, Chaker
    Inoubli, Wissem
    [J]. KNOWLEDGE AND INFORMATION SYSTEMS, 2024, 66 (09) : 5479 - 5499
  • [4] Graph Representation Learning for Large-Scale Neuronal Morphological Analysis
    Zhao, Jie
    Chen, Xuejin
    Xiong, Zhiwei
    Zha, Zheng-Jun
    Wu, Feng
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 35 (04) : 5473 - 5487
  • [5] A Compression Hashing Scheme for Large-scale Face Retrieval
    Li, Jiayong
    Ng, Wing W. Y.
    Tian, Xing
    [J]. 2018 8TH INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND TECHNOLOGY (ICIST 2018), 2018, : 245 - 251
  • [6] Large-Scale Graph Alignment Based on Topological Structure Representation Learning
    Wang, Chen-Xu
    Zhou, Jun-Ming
    Jiang, Pei-Jing
    [J]. Jisuanji Xuebao/Chinese Journal of Computers, 2023, 46 (07): : 1350 - 1365
  • [7] Large-Scale Unsupervised Hashing with Shared Structure Learning
    Liu, Xianglong
    Mu, Yadong
    Zhang, Danchen
    Lang, Bo
    Li, Xuelong
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2015, 45 (09) : 1811 - 1822
  • [8] Large-Scale Video Hashing via Structure Learning
    Ye, Guangnan
    Liu, Dong
    Wang, Jun
    Chang, Shih-Fu
    [J]. 2013 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2013, : 2272 - 2279
  • [9] GNNVis: Visualize Large-Scale Data by Learning a Graph Neural Network Representation
    Huang, Yajun
    Zhang, Jingbin
    Yang, Yiyang
    Gong, Zhiguo
    Hao, Zhifeng
    [J]. CIKM '20: PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT, 2020, : 545 - 554
  • [10] Accelerating Large-Scale Heterogeneous Interaction Graph Embedding Learning via Importance Sampling
    Ji, Yugang
    Yin, Mingyang
    Yang, Hongxia
    Zhou, Jingren
    Zheng, Vincent W.
    Shi, Chuan
    Fang, Yuan
    [J]. ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2021, 15 (01)