SILO



The weekly SILO Seminar Series is made possible through the generous support of the 3M Company and its Advanced Technology Group

3M

with additional support from the Analytics Group of the Northwestern Mutual Life Insurance Company

Northwestern Mutual

Learning Large-Scale Poisson DAG Models based on OverDispersion Scoring

Gunwoong Park, Graduate student, Department of Statistics, UW-Madison

Date and Time: Sep 09, 2015 (12:30 PM)
Location: Orchard room (3280) at the Wisconsin Institute for Discovery Building

Abstract:

We address the question of identifiability and learning algorithms for large-scale Poisson Directed Acyclic Graphical (DAG) models. We define general Poisson DAG models as models where each node is a Poisson random variable with rate parameter depending on the values of the parents in the underlying DAG. First, we prove that Poisson DAG models are identifiable from observational data, and present a polynomial-time algorithm that learns the Poisson DAG model under suitable regularity conditions. The main idea behind our algorithm is based on overdispersion, in that variables that are conditionally Poisson are overdispersed relative to variables that are marginally Poisson. Exploiting overdispersion allows
us to learn the causal ordering and then use ideas from learning large-scale regression models to reduce computational complexity. We provide both theoretical guarantees and simulation results for both small and large-scale DAGs to validate
the success of our algorithm.