Fitting block models with covariates:”Likelihood” methods ,sparsity and selection of the number of blocks

Add to your list(s) Download to your calendar using vCal

• Peter Bickel (University of California, Berkeley)
• Thursday 14 July 2016, 12:00-12:30
• Seminar Room 1, Newton Institute.

If you have a question about this talk, please contact INI IT.

This talk has been canceled/deleted

Co-authors: Purna Sarkar (U. of Texas,Austin), Soumendu Mukherjee (U C, Berkeley), Sharmodeep Bhattacharyya, (Oregon State University)

We introduce block models with edge and block covariates ,along the lines of Hoff, Handcock,Raftery (2002) ,specializing to covariate forms of the types proposed by Zhao, Levina,Zhu(2014) and generalizing that of Newman ,Clauset(2015)..WE study mean field variational fitting in this situation along the lines of Celisse,Daudin,Pierre( 2011 ) and B.,Choi,Chang,Zhang (2013).Using theory,examples, and simulations we argue that these methods ,possibly regularized lead to models with a sparse number of blocks. 

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

This talk is not included in any other list

Note that ex-directory lists are not shown.