Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
The paper introduces a notion of low rank conditional probability tables (CPTs). and presents a method for learning Bayesian network structures with low rank CPTs. Overall the reviewers found the results innovative, interesting and theoretically justified. Many of the reviewers concerns were properly addressed in the rebuttal. Several reviewers stated that more empirical work could have greatly benefit the paper.