Invited Speakers

Ceren Budak, Microsoft Research - How valuable are social graphs in predicting adoption behavior?
The use of cascade models and online social networks data to study adoption behavior is common practice, both among the research community and the industry. But how valuable are social graphs, or at least the ones we have access to, for modeling adoption behavior? And how does this value vary across different networks? In this talk, I will tackle these questions, challenge the common use of local influence models and revive theories introduced in diffusion of innovations research to model human behavior.

Pierre-Andre Maugis, University College London - Network Analysis and Nonparametric Subgraph Count Statistics
Networks are ubiquitous in today’s world. Any time we make observations about people, places, or things and the interactions between them, we have a network. Yet a quantitative understanding of real-world networks is in its infancy, and must be based on strong theoretical and methodological foundations. The goal of this talk is to provide some insight into these foundations from the perspective of nonparametric techniques in relation to subgraph count statistics, in particular how tradeoffs between model complexity and parsimony can be balanced to yield practical algorithms with provable properties.

Dena Asta, Carnegie Mellon - Geometric Network Comparison
An overriding theme in this talk is that an understanding of the relevant geometric structure in a network is useful for its efficient and large-scale statistical analysis. I will discuss a geometric approach to network analysis that approximates networks as probability distributions on negatively curved manifolds. I will then propose a more general, principled statistical approach to network comparison, based on the non-parametric inference and comparison of densities on hyperbolic manifolds from sample networks. This talk represents joint work with Cosma Shalizi.