{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T18:01:07Z","timestamp":1791396067521,"version":"4.3.3"},"reference-count":43,"publisher":"Institute of Mathematical Statistics","issue":"2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Bayesian Anal."],"published-print":{"date-parts":[[2024,6,1]]},"abstract":"<jats:p>The exponential random graph model (ERGM) is a popular model for social networks, which is known to have an intractable likelihood function. Sampling from the posterior for such a model is a long-standing problem in statistical research. We analyze the performance of the stochastic gradient Langevin dynamics (SGLD) algorithm (also known as noisy Longevin Monte Carlo) in tackling this problem, where the stochastic gradient is calculated via running a short Markov chain (the so-called inner Markov chain in this paper) at each iteration. We show that if the model size grows with the network size slowly enough, then SGLD converges to the true posterior in 2-Wasserstein distance as the network size and iteration number become large regardless of the length of the inner Markov chain performed at each iteration. Our study provides a scalable algorithm for analyzing large-scale social networks with possibly high-dimensional ERGMs.<\/jats:p>","DOI":"10.1214\/23-ba1364","type":"journal-article","created":{"date-parts":[[2023,1,23]],"date-time":"2023-01-23T07:37:35Z","timestamp":1674459455000},"page":"595-621","source":"Crossref","is-referenced-by-count":2,"title":["Bayesian Analysis of Exponential Random Graph Models Using Stochastic Gradient Markov Chain Monte Carlo"],"prefix":"10.1214","volume":"19","author":[{"given":"Qian","family":"Zhang","sequence":"first","affiliation":[{"name":"Department of Statistics, Purdue University, West Lafayette, IN 47907, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Faming","family":"Liang","sequence":"additional","affiliation":[{"name":"Department of Statistics, Purdue University, West Lafayette, IN 47907, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"108","reference":[{"key":"2","doi-asserted-by":"publisher","unstructured":"Atchade, Y. F., Lartillot, N., and Robert, C. P. (2013). \u201cBayesian computation for intractable normalizing constants.\u201d <i>Brazilian Journal of Statistics<\/i>, 27: 416\u2013436.","DOI":"10.1214\/11-BJPS174"},{"key":"4","doi-asserted-by":"publisher","unstructured":"Besag, J. (1974). \u201cSpatial interaction and the statistical analysis of lattice systems.\u201d <i>Journal of the Royal Statistical Society: Series B (Methodological)<\/i>, 36(2): 192\u2013225.","DOI":"10.1111\/j.2517-6161.1974.tb00999.x"},{"key":"5","doi-asserted-by":"publisher","unstructured":"Besag, J. E. and Moran, P. A. (1975). \u201cOn the estimation and testing of spatial interaction in Gaussian lattice processes.\u201d <i>Biometrika<\/i>, 62(3): 555\u2013562.","DOI":"10.1093\/biomet\/62.3.555"},{"key":"6","unstructured":"Bhatia, K., Ma, Y.-A., Dragan, A. D., Bartlett, P. L., and Jordan, M. I. (2019). \u201cBayesian Robustness: A Nonasymptotic Viewpoint.\u201d <i>arXiv preprint arXiv:1907.11826<\/i>."},{"key":"7","doi-asserted-by":"crossref","unstructured":"Caimo, A. and Friel, N. (2011). \u201cBayesian inference for exponential random graph models.\u201d <i>Social Networks<\/i>, 33: 41\u201355.","DOI":"10.1016\/j.socnet.2010.09.004"},{"key":"8","doi-asserted-by":"crossref","unstructured":"Childs, A. M., Patterson, R. B., and MacKay, D. J. (2001). \u201cExact sampling from non-attractive distributions using summary states.\u201d <i>Physics Review E<\/i>, 63: 036113.","DOI":"10.1103\/PhysRevE.63.036113"},{"key":"9","doi-asserted-by":"publisher","unstructured":"Dalalyan, A. S. and Karagulyan, A. (2019). \u201cUser-friendly guarantees for the Langevin Monte Carlo with inaccurate gradient.\u201d <i>Stochastic Processes and their Applications<\/i>, 129(12): 5278 \u2013 5311. URL <a href=\"http:\/\/www.sciencedirect.com\/science\/article\/pii\/S0304414918304824\">http:\/\/www.sciencedirect.com\/science\/article\/pii\/S0304414918304824<\/a>","DOI":"10.1016\/j.spa.2019.02.016"},{"key":"10","unstructured":"Durmus, A., Majewski, S., and Miasojedow, B. (2019). \u201cAnalysis of Langevin Monte Carlo via Convex Optimization.\u201d <i>J. Mach. Learn. Res.<\/i>, 20: 73:1\u201373:46."},{"key":"11","doi-asserted-by":"publisher","unstructured":"Everitt, R. G. (2012). \u201cBayesian parameter estimation for latent Markov random fields and social networks.\u201d <i>Journal of Computational and Graphical Statistics<\/i>, 21: 940\u2013960.","DOI":"10.1080\/10618600.2012.687493"},{"key":"12","unstructured":"Fellows, I. and Handcock, M. (2017). \u201cRemoving Phase Transitions from Gibbs Measures.\u201d In Singh, A. and Zhu, J. (eds.), <i>Proceedings of the 20th International Conference on Artificial Intelligence and Statistics<\/i>, volume 54 of <i>Proceedings of Machine Learning Research<\/i>, 289\u2013297. Fort Lauderdale, FL, USA: PMLR."},{"key":"13","doi-asserted-by":"crossref","unstructured":"Geman, S. and Geman, D. (1984). \u201cStochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images.\u201d <i>IEEE Transactions on Pattern Analysis and Machine Intelligence<\/i>, PAMI-6: 721\u2013741.","DOI":"10.1109\/TPAMI.1984.4767596"},{"key":"14","doi-asserted-by":"crossref","unstructured":"Geyer, C. J. and Thompson, E. A. (1992). \u201cConstrained Monte Carlo Maximum Likelihood for Dependent Data.\u201d <i>Journal of the Royal Statistical Society. Series B (Methodological)<\/i>, 54(3): 657\u2013699. URL <a href=\"http:\/\/www.jstor.org\/stable\/2345852\">http:\/\/www.jstor.org\/stable\/2345852<\/a>","DOI":"10.1111\/j.2517-6161.1992.tb01443.x"},{"key":"15","doi-asserted-by":"crossref","unstructured":"Gibbs, A. and Su, F. (2002). \u201cOn Choosing and Bounding Probability Metrics.\u201d <i>International Statistical Review<\/i>, 70(3): 419\u2013435.","DOI":"10.1111\/j.1751-5823.2002.tb00178.x"},{"key":"16","doi-asserted-by":"publisher","unstructured":"Goldstein, J., Haran, M., Simeonov, I., Fricks, J., and Chiaromonte, F. (2015). \u201cAn attraction\u2013repulsion point process model for respiratory syncytial virus infections.\u201d <i>Biometrics<\/i>, 71(2): 376\u2013385.","DOI":"10.1111\/biom.12267"},{"key":"17","doi-asserted-by":"publisher","unstructured":"Hunter, D. R., Goodreau, S. M., and Handcock, M. S. (2008). \u201cGoodness of fit of social network models.\u201d <i>Journal of the American Statistical Association<\/i>, 103: 248\u2013258.","DOI":"10.1198\/016214507000000446"},{"key":"18","doi-asserted-by":"publisher","unstructured":"Jin, I. H. and Liang, F. (2013). \u201cFitting social network models using varying truncation stochastic approximation MCMC algorithm.\u201d <i>Journal of computational and graphical statistics<\/i>, 22(4): 927\u2013952.","DOI":"10.1080\/10618600.2012.680851"},{"key":"19","doi-asserted-by":"publisher","unstructured":"Jin, I. H. and Liang, F. (2014). \u201cUse of SAMC for Bayesian Analysis of Statistical Models with Intractable Normalizing Constants.\u201d <i>Computational Statistics and Data Analysis<\/i>, 71: 402\u2013416.","DOI":"10.1016\/j.csda.2012.07.005"},{"key":"20","unstructured":"Kapferer, B. (1972). <i>Strategy and transaction in an African factory<\/i>. Manchester: Manchester University Press."},{"key":"21","doi-asserted-by":"publisher","unstructured":"Kim, B., Lee, K. H., Xue, L., and Niu, X. (2018). \u201cA review of dynamic network models with latent variables.\u201d <i>Statistics surveys<\/i>, 12: 105.","DOI":"10.1214\/18-SS121"},{"key":"22","doi-asserted-by":"publisher","unstructured":"Liang, F. (2007). \u201cContinuous contour Monte Carlo for marginal density estimation with an application to a spatial statistical models.\u201d <i>Journal of Computational and Graphical Statistics<\/i>, 16: 608\u2013632.","DOI":"10.1198\/106186007X238459"},{"key":"23","doi-asserted-by":"publisher","unstructured":"Liang, F. (2010). \u201cA double Metropolis-Hastings sampler for spatial models with intractable normalizing constants.\u201d <i>Journal of Statistical Computing and Simulation<\/i>, 80: 1007\u20131022.","DOI":"10.1080\/00949650902882162"},{"key":"24","doi-asserted-by":"publisher","unstructured":"Liang, F. and Jin, I. H. (2013). \u201cA Monte Carlo Metropolis-Hastings Algorithm for Sampling from Distributions with Intractable Normalizing Constants.\u201d <i>Neural Computation<\/i>, 25: 2199\u20132234.","DOI":"10.1162\/NECO_a_00466"},{"key":"28","doi-asserted-by":"crossref","unstructured":"Morris, M., Handcock, M., and Hunter, D. (2008). \u201cSpecification of Exponential-Family Random Graph Models: Terms and Computational Aspects.\u201d <i>Journal of statistical software<\/i>, 24 4: 1548\u20137660.","DOI":"10.18637\/jss.v024.i04"},{"key":"29","unstructured":"Murray, I., Ghahramani, Z., and MacKay, D. J. (2006). \u201cMCMC for doubly-intractable distributions.\u201d In <i>Proceedings of the Twenty-Second Conference on Uncertainty in Artificial Intelligence (UAI)<\/i>, 359\u2013366."},{"key":"30","unstructured":"Padgett, J. (1994). <i>Marriage and elite structure in Renaissance Florence, 1282\u20131500<\/i>. Social Science History Association."},{"key":"31","doi-asserted-by":"publisher","unstructured":"Park, J. and Haran, M. (2018). \u201cBayesian inference in the presence of intractable normalizing functions.\u201d <i>Journal of the American Statistical Association<\/i>, 113(523): 1372\u20131390.","DOI":"10.1080\/01621459.2018.1448824"},{"key":"33","unstructured":"Riggan, W. B., Creason, J. P., Nelson, W. C., Manton, K. G., Woodbury, M. A., Stallard, E., Pellom, A. C., and Beaubier, J. (1987). <i>U.S. Cancer Mortality Rates and Trends, 1950\u20131979. (Vol. IV: Maps)<\/i>. U.S. Government Printing Office.: U.S. Government Printing Office."},{"key":"35","doi-asserted-by":"crossref","unstructured":"Robins, G., Pattison, P., Kalish, Y., and Lusher, D. (2007a). \u201cAn introduction to exponential random graph (p*) models for social networks.\u201d <i>Social Networks<\/i>, 29(2): 173\u2013191. Special Section: Advances in Exponential Random Graph (p*) Models. URL <a href=\"http:\/\/www.sciencedirect.com\/science\/article\/pii\/S0378873306000372\">http:\/\/www.sciencedirect.com\/science\/article\/pii\/S0378873306000372<\/a>","DOI":"10.1016\/j.socnet.2006.08.002"},{"key":"36","doi-asserted-by":"crossref","unstructured":"Robins, G. E., Snijers, T. A. B., Wang, P., Handcock, M. S., and Pattison, P. E. (2007b). \u201cRecent development in exponential random graph models for social networks.\u201d <i>Social Networks<\/i>, 29: 192\u2013215.","DOI":"10.1016\/j.socnet.2006.08.003"},{"key":"37","doi-asserted-by":"crossref","unstructured":"Rozemberczki, B. and Sarkar, R. (2020). \u201cCharacteristic Functions on Graphs: Birds of a Feather, from Statistical Descriptors to Parametric Models.\u201d In <i>Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM \u201920)<\/i>, 1325\u20131334. ACM.","DOI":"10.1145\/3340531.3411866"},{"key":"38","doi-asserted-by":"publisher","unstructured":"Salter-Townshend, M. and Brendan Murphy, T. (2015). \u201cRole analysis in networks using mixtures of exponential random graph models.\u201d <i>Journal of Computational and Graphical Statistics<\/i>, 24(2): 520\u2013538.","DOI":"10.1080\/10618600.2014.923777"},{"key":"39","doi-asserted-by":"publisher","unstructured":"Song, Q., Sun, Y., Ye, M., and Liang, F. (2020). \u201cExtended stochastic gradient Markov chain Monte Carlo for large-scale Bayesian variable selection.\u201d <i>Biometrika<\/i>, in press.","DOI":"10.1093\/biomet\/asaa029"},{"key":"40","doi-asserted-by":"publisher","unstructured":"Stoehr, J., Benson, A., and Friel, N. (2019). \u201cNoisy Hamiltonian Monte Carlo for doubly intractable distributions.\u201d <i>Journal of Computational and Graphical Statistics<\/i>, 28(1): 220\u2013232.","DOI":"10.1080\/10618600.2018.1506346"},{"key":"41","unstructured":"Teh, W., Thiery, A., and Vollmer, S. (2016). \u201cConsistency and fluctuations for stochastic gradient Langevin dynamics.\u201d <i>Journal of Machine Learning Research<\/i>, 17: 1\u201333."},{"key":"43","doi-asserted-by":"publisher","unstructured":"Zhang, Q. and Liang, F. (2023). \u201cSupplementary Material for \u201c Bayesian analysis of exponential random graph models using stochastic gradient Markov chain Monte Carlo\u201d.\u201d <i>Bayesian Analysis<\/i>.","DOI":"10.1214\/23-BA1364SUPP"},{"key":"1","doi-asserted-by":"publisher","unstructured":"Alquier, P., Friel, N., Everitt, R., and Boland, A. (2016). \u201cNoisy Monte Carlo: Convergence of Markov chains with approximate transition kernels.\u201d <i>Statistics and Computing<\/i>, 26(1-2): 29\u201347.","DOI":"10.1007\/s11222-014-9521-x"},{"key":"27","doi-asserted-by":"publisher","unstructured":"M\u00f8ller, J., Pettitt, A. N., Reeves, R., and Berthelsen, K. K. (2006). \u201cAn efficient Markov chain Monte Carlo method for distributions with intractable normalising constants.\u201d <i>Biometrika<\/i>, 93(2): 451\u2013458.","DOI":"10.1093\/biomet\/93.2.451"},{"key":"34","doi-asserted-by":"publisher","unstructured":"Robbins, H. and Monro, S. (1951). \u201cA stochastic approximation method.\u201d <i>The Annals of Mathematical Statistics<\/i>, 22(3): 400\u2013407.","DOI":"10.1214\/aoms\/1177729586"},{"key":"3","doi-asserted-by":"publisher","unstructured":"Barbieri, M. and Berger, J. (2004). \u201cOptimal predictive model selection.\u201d <i>The Annals of Statistics<\/i>, 32: 870\u2013897.","DOI":"10.1214\/009053604000000238"},{"key":"42","unstructured":"Welling, M. and Teh, Y. W. (2011). \u201cBayesian learning via stochastic gradient Langevin dynamics.\u201d In <i>Proceedings of the 28th International Conference on Machine Learning (ICML-11)<\/i>, 681\u2013688."},{"key":"25","doi-asserted-by":"publisher","unstructured":"Liang, F., Jin, I. H., Song, Q., and Liu, J. S. (2016). \u201cAn Adaptive Exchange Algorithm for Sampling from Distributions with Intractable Normalizing Constants.\u201d <i>Journal of the American Statistical Association<\/i>, 111(513): 377\u2013393.","DOI":"10.1080\/01621459.2015.1009072"},{"key":"26","doi-asserted-by":"publisher","unstructured":"Liang, F., Song, Q., and Yu, K. (2013). \u201cBayesian subset modeling for high-dimensional generalized linear models.\u201d <i>Journal of the American Statistical Association<\/i>, 108(502): 589\u2013606.","DOI":"10.1080\/01621459.2012.761942"},{"key":"32","doi-asserted-by":"publisher","unstructured":"Propp, J. G. and Wilson, D. B. (1996). \u201cExact sampling with coupled Markov chains and applications to statistical mechanics.\u201d <i>Random Structures and Algorithms<\/i>, 9: 223\u2013252.","DOI":"10.1002\/(SICI)1098-2418(199608\/09)9:1\/2<223::AID-RSA14>3.3.CO;2-R"}],"container-title":["Bayesian Analysis"],"original-title":[],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:39:21Z","timestamp":1791394761000},"score":1,"resource":{"primary":{"URL":"https:\/\/projecteuclid.org\/journals\/bayesian-analysis\/volume-19\/issue-2\/Bayesian-Analysis-of-Exponential-Random-Graph-Models-Using-Stochastic-Gradient\/10.1214\/23-BA1364.full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,6,1]]},"references-count":43,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2024,6,1]]}},"URL":"https:\/\/doi.org\/10.1214\/23-ba1364","relation":{},"ISSN":["1936-0975","1931-6690"],"issn-type":[{"value":"1936-0975","type":"print"},{"value":"1931-6690","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,6,1]]}}}