On Mixing in Pairwise Markov Random Fields with Application to Social Networks

被引:1
作者
Avrachenkov, Konstantin [1 ]
Iskhakov, Lenar [2 ]
Mironov, Maksim [2 ]
机构
[1] Inria Sophia Antipolis, 2004 Route Lucioles, Sophia Antipolis, France
[2] Moscow Inst Phys & Technol, Dolgoprudnyi, Russia
来源
ALGORITHMS AND MODELS FOR THE WEB GRAPH, WAW 2016 | 2016年 / 10088卷
关键词
ENERGY MINIMIZATION; MODELS;
D O I
10.1007/978-3-319-49787-7_11
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We consider pairwise Markov random fields which have a number of important applications in statistical physics, image processing and machine learning such as Ising model and labeling problem to name a couple. Our own motivation comes from the need to produce synthetic models for social networks with attributes. First, we give conditions for rapid mixing of the associated Glauber dynamics and consider interesting particular cases. Then, for pairwise Markov random fields with submodular energy functions we construct monotone perfect simulation.
引用
收藏
页码:127 / 139
页数:13
相关论文
共 18 条
  • [1] [Anonymous], 1998, ACM SIGMOD RECORD, DOI DOI 10.1145/276305.276332
  • [2] [Anonymous], 2009, American Mathematical Soc.
  • [3] [Anonymous], 2004, P 10 ACM SIGKDD INT, DOI DOI 10.1145/1014052.1014062
  • [4] Subsampling for Chain-Referral Methods
    Avrachenkov, Konstantin
    Neglia, Giovanni
    Tuholukova, Alina
    [J]. ANALYTICAL AND STOCHASTIC MODELLING TECHNIQUES AND APPLICATIONS, 2016, 9845 : 17 - 31
  • [5] Markov random fields with efficient approximations
    Boykov, Y
    Veksler, O
    Zabih, R
    [J]. 1998 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, 1998, : 648 - 655
  • [6] Fast approximate energy minimization via graph cuts
    Boykov, Y
    Veksler, O
    Zabih, R
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2001, 23 (11) : 1222 - 1239
  • [7] Bremaud Pierre., 1998, Texts in Applied Mathematics, V31
  • [8] Freeman L. C., SOCIAL NETWORKS DATA
  • [9] Report on the theory of ferromagnetism
    Ising, E
    [J]. ZEITSCHRIFT FUR PHYSIK, 1925, 31 : 253 - 258
  • [10] Approximation algorithms for classification problems with pairwise relationships:: Metric labeling and Markov random fields
    Kleinberg, J
    Tardos, É
    [J]. JOURNAL OF THE ACM, 2002, 49 (05) : 616 - 639