Bayesian Inference for low-rank Ising networks

12 maart 2015
Auteur: Marsman, M., Maris, G., Bechger, T. M., & Glas, C. A. W.

Estimating the structure of Ising networks is a notoriously difficult problem. In this article it is demonstrated that using a latent variable representation of the Ising network, a full-data-information approach to uncover the network structure can be employed. Thereby, only ignoring information encoded in the prior distribution (of the latent variables). The full-data-information approach avoids having to compute the partition function and is thus computationally feasible, even for networks with many nodes. The full-data-information approach is illustrated with the estimation of dense networks.

Ga naar maakt gebruik van cookies, onder andere om de website te analyseren en het gebruiksgemak te vergroten. Via de knop ‘instellen’ geef je aan welke cookies je wilt accepteren. Meer informatie over Cookies en privacy