Distance metric learning for graph structured data

被引:0
|
作者
Tomoki Yoshida
Ichiro Takeuchi
Masayuki Karasuyama
机构
[1] Nagoya Institute of Technology,
[2] National Institute for Material Science,undefined
[3] RIKEN Center for Advanced Intelligence Project,undefined
[4] Japan Science and Technology Agency,undefined
来源
Machine Learning | 2021年 / 110卷
关键词
Metric learning; Structured data; Graph mining; Convex optimization; Interpretability;
D O I
暂无
中图分类号
学科分类号
摘要
Graphs are versatile tools for representing structured data. As a result, a variety of machine learning methods have been studied for graph data analysis. Although many such learning methods depend on the measurement of differences between input graphs, defining an appropriate distance metric for graphs remains a controversial issue. Hence, we propose a supervised distance metric learning method for the graph classification problem. Our method, named interpretable graph metric learning (IGML), learns discriminative metrics in a subgraph-based feature space, which has a strong graph representation capability. By introducing a sparsity-inducing penalty on the weight of each subgraph, IGML can identify a small number of important subgraphs that can provide insight into the given classification task. Because our formulation has a large number of optimization variables, an efficient algorithm that uses pruning techniques based on safe screening and working set selection methods is also proposed. An important property of IGML is that solution optimality is guaranteed because the problem is formulated as a convex problem and our pruning strategies only discard unnecessary subgraphs. Furthermore, we show that IGML is also applicable to other structured data such as itemset and sequence data, and that it can incorporate vertex-label similarity by using a transportation-based subgraph feature. We empirically evaluate the computational efficiency and classification performance of IGML on several benchmark datasets and provide some illustrative examples of how IGML identifies important subgraphs from a given graph dataset.
引用
收藏
页码:1765 / 1811
页数:46
相关论文
共 50 条
  • [1] Distance metric learning for graph structured data
    Yoshida, Tomoki
    Takeuchi, Ichiro
    Karasuyama, Masayuki
    MACHINE LEARNING, 2021, 110 (07) : 1765 - 1811
  • [2] Learning to classify structured data by graph propositionalization
    Karunaratne, Thashmee
    Bostrom, Henrik
    PROCEEDINGS OF THE SECOND IASTED INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE, 2006, : 393 - +
  • [3] Distance metric learning by minimal distance maximization
    Yu, Yaoliang
    Jiang, Jiayan
    Zhang, Liming
    PATTERN RECOGNITION, 2011, 44 (03) : 639 - 649
  • [4] Distance Metric Learning with Eigenvalue Optimization
    Ying, Yiming
    Li, Peng
    JOURNAL OF MACHINE LEARNING RESEARCH, 2012, 13 : 1 - 26
  • [5] Globality and locality incorporation in distance metric learning
    Wang, Wei
    Hu, Bao-Gang
    Wang, Zeng-Fu
    NEUROCOMPUTING, 2014, 129 : 185 - 198
  • [6] Safe Triplet Screening for Distance Metric Learning
    Yoshida, Tomoki
    Takeuchi, Ichiro
    Karasuyama, Masayuki
    KDD'18: PROCEEDINGS OF THE 24TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2018, : 2653 - 2662
  • [7] Efficient Dual Approach to Distance Metric Learning
    Shen, Chunhua
    Kim, Junae
    Liu, Fayao
    Wang, Lei
    van den Hengel, Anton
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2014, 25 (02) : 394 - 406
  • [8] Distance metric learning with the Universum
    Bac Nguyen
    Morell, Carlos
    De Baets, Bernard
    PATTERN RECOGNITION LETTERS, 2017, 100 : 37 - 43
  • [9] Learning Interpretable Metric between Graphs: Convex Formulation and Computation with Graph Mining
    Yoshida, Tomoki
    Takeuchi, Ichiro
    Karasuyama, Masayuki
    KDD'19: PROCEEDINGS OF THE 25TH ACM SIGKDD INTERNATIONAL CONFERENCCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2019, : 1026 - 1036
  • [10] Graph Representation Learning With Adaptive Metric
    Zhang, Chun-Yang
    Cai, Hai-Chun
    Chen, C. L. Philip
    Lin, Yue-Na
    Fang, Wu-Peng
    IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING, 2023, 10 (04): : 2074 - 2085