Conference paper Open Access

Dependent Relational Gamma Process Models for Longitudinal Networks

Yang, Sikun; Koeppl, Heinz

A probabilistic framework based on the covariate-dependent relational gamma process is developed to analyze relational data arising from longitudinal networks. The proposed framework characterizes networked nodes by nonnegative node-group memberships, which allow each node to belong to multiple latent groups simultaneously, and encodes edge probabilities between each pair of nodes using a Bernoulli Poisson link to the embedded latent space. Within the latent space, our framework models the birth and death dynamics of individual groups via a thinning function. Our framework also captures the evolution of individual
node-group memberships over time using gamma Markov processes. Exploiting the recent advances in data augmentation and marginalization techniques, a simple and efficient Gibbs sampler is proposed for posterior computation. Experimental
results on a simulation study and three real world temporal network data sets demonstrate the model’s capability, competitive performance and scalability compared to state-of-the-art methods.

Files (1.7 MB)
Name Size
1.7 MB Download
All versions This version
Views 8888
Downloads 4444
Data volume 73.0 MB73.0 MB
Unique views 8585
Unique downloads 4343


Cite as