Wasserstein-Based Graph Alignment

被引:7
|
作者
Maretic, Hermina Petric [1 ,2 ]
El Gheche, Mireille [3 ]
Minder, Matthias [1 ]
Chierchia, Giovanni [4 ]
Frossard, Pascal [1 ]
机构
[1] Ecole Polytech Fed Lausanne, CH-1015 Lausanne, Switzerland
[2] Amazon, 1 Principal Pl,Worship St, London EC2A 2FA, England
[3] Sony Europe BV, Sony AI, CH-8952 Schlieren, Switzerland
[4] Univ Gustave Eiffel, CNRS, UMR 8049, LIGM,ESIEE Paris, F-93162 Noisy Le Grand, France
来源
IEEE TRANSACTIONS ON SIGNAL AND INFORMATION PROCESSING OVER NETWORKS | 2022年 / 8卷
关键词
Graph alignment; graph comparison; graph distance; optimal transport; wasserstein distance; ALGORITHM;
D O I
10.1109/TSIPN.2022.3169632
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
A novel method for comparing non-aligned graphs of various sizes is proposed, based on the Wasserstein distance between graph signal distributions induced by the respective graph Laplacian matrices. Specifically, a new formulation for the one-to-many graph alignment problem is casted, which aims at matching a node in the smaller graph with one or more nodes in the larger graph. By incorporating optimal transport into our graph comparison framework, a structurally-meaningful graph distance, and a signal transportation plan that models the structure of graph data are generated. The resulting alignment problem is solved with stochastic gradient descent, where a novel Dykstra operator is used to ensure that the solution is a one-to-many (soft) assignment matrix. The performance of our novel framework is demonstrated on graph alignment, graph classification and graph signal transportation. Our method is shown to lead to significant improvements with respect to the state-of-the-art algorithms on each ofthese tasks.
引用
收藏
页码:353 / 363
页数:11
相关论文
共 50 条
  • [1] Unsupervised Graph Alignment with Wasserstein Distance Discriminator
    Gao, Ji
    Huang, Xiao
    Li, Jundong
    KDD '21: PROCEEDINGS OF THE 27TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2021, : 426 - 435
  • [2] WOOD: Wasserstein-Based Out-of-Distribution Detection
    Wang, Yinan
    Sun, Wenbo
    Jin, Jionghua
    Kong, Zhenyu
    Yue, Xiaowei
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2024, 46 (02) : 944 - 956
  • [3] Dynamic clustering of interval data using a Wasserstein-based distance
    Irpino, Antonio
    Verde, Rosanna
    PATTERN RECOGNITION LETTERS, 2008, 29 (11) : 1648 - 1658
  • [4] Wasserstein-based fairness interpretability framework for machine learning models
    Alexey Miroshnikov
    Konstandinos Kotsiopoulos
    Ryan Franks
    Arjun Ravi Kannan
    Machine Learning, 2022, 111 : 3307 - 3357
  • [5] Wasserstein-based fairness interpretability framework for machine learning models
    Miroshnikov, Alexey
    Kotsiopoulos, Konstandinos
    Franks, Ryan
    Kannan, Arjun Ravi
    MACHINE LEARNING, 2022, 111 (09) : 3307 - 3357
  • [6] Personalized Purchase Prediction of Market Baskets with Wasserstein-Based Sequence Matching
    Kraus, Mathias
    Feuerriegel, Stefan
    KDD'19: PROCEEDINGS OF THE 25TH ACM SIGKDD INTERNATIONAL CONFERENCCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2019, : 2643 - 2652
  • [7] Wasserstein-Based Evolutionary Operators for Optimizing Sets of Points: Application to Wind-Farm Layout Design
    Sow, Babacar
    Le Riche, Rodolphe
    Pelamatti, Julien
    Keller, Merlin
    Zannane, Sanaa
    APPLIED SCIENCES-BASEL, 2024, 14 (17):
  • [8] Persistent homology based Wasserstein distance for graph networks
    Babu, Archana
    John, Sunil Jacob
    HACETTEPE JOURNAL OF MATHEMATICS AND STATISTICS, 2025, 54 (01): : 90 - 114
  • [9] LCS graph kernel based on Wasserstein distance in longest common subsequence metric space
    Huang, Jianming
    Fang, Zhongxi
    Kasai, Hiroyuki
    SIGNAL PROCESSING, 2021, 189
  • [10] A Novel Graph Kernel Based on the Wasserstein Distance and Spectral Signatures
    Liu, Yantao
    Rossi, Luca
    Torsello, Andrea
    STRUCTURAL, SYNTACTIC, AND STATISTICAL PATTERN RECOGNITION, S+SSPR 2022, 2022, 13813 : 122 - 131