A Markov Chain Monte Carlo Approach for Source Detection in Networks

被引:3
作者
Zhang, Le [1 ]
Jin, Tianyuan [1 ]
Xu, Tong [1 ]
Chang, Biao [1 ]
Wang, Zhefeng [1 ]
Chen, Enhong [1 ]
机构
[1] Univ Sci & Technol China, Sch Comp Sci & Technol, Anhui Prov Key Lab Big Data Anal & Applicat, Hefei, Anhui, Peoples R China
来源
SOCIAL MEDIA PROCESSING, SMP 2017 | 2017年 / 774卷
基金
中国国家自然科学基金; 美国国家科学基金会;
关键词
Social network; Source detection; Markov Chain Monte Carlo;
D O I
10.1007/978-981-10-6805-8_7
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The source detection task, which targets at finding the most likely source given a snapshot of the information diffusion, has attracted wide attention in theory and practice. However, due to the hardness of this task, traditional techniques may suffer biased solution and extraordinary time complexity. Specially, source detection task based on the widely used Linear Threshold (LT) model has been largely ignored. To that end, in this paper, we formulate the source detection task as a Maximum Likelihood Estimation (MLE) problem, and then proposed a Markov Chain Monte Carlo (MCMC) algorithm, whose convergence is demonstrated. Along this line, to further improve efficiency of proposed algorithm, the sampling method is simplified only on the observed graph rather than the entire one. Extensive experiments on public data sets show that our MCMC algorithm significantly outperforms several state-of- the-art baselines, which validates the potential of our algorithm in source detection task.
引用
收藏
页码:77 / 88
页数:12
相关论文
共 50 条
[41]   A multiobjective Markov chain Monte Carlo approach for history matching and uncertainty quantification [J].
Olalotiti-Lawal, Feyi ;
Datta-Gupta, Akhil .
JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING, 2018, 166 :759-777
[42]   Regenerative Markov Chain Monte Carlo for Any Distribution [J].
Minh, Do Le ;
Minh, David D. L. ;
Nguyen, Andrew L. .
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2012, 41 (09) :1745-1760
[43]   LOCAL DEGENERACY OF MARKOV CHAIN MONTE CARLO METHODS [J].
Kamatani, Kengo .
ESAIM-PROBABILITY AND STATISTICS, 2014, 18 :713-725
[44]   Markov chain Monte Carlo1. Examples [J].
Arnab Chakraborty .
Resonance, 2002, 7 (3) :25-34
[45]   SEQUENTIALLY INTERACTING MARKOV CHAIN MONTE CARLO METHODS [J].
Brockwell, Anthony ;
Del Moral, Pierre ;
Doucet, Arnaud .
ANNALS OF STATISTICS, 2010, 38 (06) :3387-3411
[46]   Bayesian Computation Via Markov Chain Monte Carlo [J].
Craiu, Radu V. ;
Rosenthal, Jeffrey S. .
ANNUAL REVIEW OF STATISTICS AND ITS APPLICATION, VOL 1, 2014, 1 :179-201
[47]   OPTIMAL VARIANCE REDUCTION FOR MARKOV CHAIN MONTE CARLO [J].
Huang, Lu-Jing ;
Liao, Yin-Ting ;
Chen, Ting-Li ;
Hwang, Chii-Ruey .
SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 2018, 56 (04) :2977-2996
[48]   Ensemble preconditioning for Markov chain Monte Carlo simulation [J].
Benedict Leimkuhler ;
Charles Matthews ;
Jonathan Weare .
Statistics and Computing, 2018, 28 :277-290
[49]   A Markov Chain Monte Carlo Algorithm for Spatial Segmentation [J].
Raveendran, Nishanthi ;
Sofronov, Georgy .
INFORMATION, 2021, 12 (02) :1-14
[50]   Statistical Robustness of Markov Chain Monte Carlo Accelerators [J].
Zhang, Xiangyu ;
Bashizade, Ramin ;
Wang, Yicheng ;
Mukherjee, Sayan ;
Lebeck, Alvin R. .
ASPLOS XXVI: TWENTY-SIXTH INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS, 2021, :959-974