Timezone: »

 
Poster
Contextual Stochastic Block Models
Yash Deshpande · Subhabrata Sen · Andrea Montanari · Elchanan Mossel

Wed Dec 05 02:00 PM -- 04:00 PM (PST) @ Room 210 #79
in Wed Poster Session B »

We provide the first information theoretical tight analysis for inference of latent community structure given a sparse graph along with high dimensional node covariates, correlated with the same latent communities. Our work bridges recent theoretical breakthroughs in detection of latent community structure without nodes covariates and a large body of empirical work using diverse heuristics for combining node covariates with graphs for inference. The tightness of our analysis implies in particular, the information theoretic necessity of combining the different sources of information. Our analysis holds for networks of large degrees as well as for a Gaussian version of the model.

Keywords: Information Theory · Graphical Models · Frequentist Statistics · Belief Propagation · Statistical Physics of Learning
[ Paper] 

Author Information

Yash Deshpande (Massachusetts Institute of Technology)
Subhabrata Sen (Massachusetts Institute of Technology)
Andrea Montanari (Stanford)
Elchanan Mossel (MIT)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors