Stochastic Blockmodels and Other Models of Biological and Information Networks

Seminar
Wednesday, November 09, 2011
6:00 PM
Free and open to the public

Observations consisting of measurements on pairs of objects (or conditions)arise in a number of settings in the biological sciences (www.yeastgenome.org), with collections of scientific publications (www.jstor.org) and other hyper-linked resources (www.wikipedia.org), and insocial networks (www.linkedin.com). Analyses of such data typically aim atidentifying structure among the units of interest, in a low dimensionalspace, to support the generation of substantive hypotheses, to partiallyautomate semantic categorization, to facilitate browsing, and to simplifycomplex data into useful patterns, more in general. In this lecture, wewill survey a few exchangeable graph models. We will then focus on thestochastic blockmodel and show its utility as a quantitative tool forexploring static/dynamic networks. Within this modeling context, we discussalternative specifications and extensions that address fundamental issuesin data analysis of complex interacting systems: bridging global and localphenomena, data integration, dynamics, and scalable inference

x x

Speaker

Edo Airoldi

Professor
Harvard University