NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:5922
Title:Bias Correction of Learned Generative Models using Likelihood-Free Importance Weighting


		
Congratulations, your paper has been accepted for publication at NeurIPS2019. The reviewers found it to be a novel well executed piece of work. When preparing the camera ready version, please bear in mind the reviewers comments. In particular - Please carefully define what bias is. The footnote on p1 is somewhat vague. Is a generative model biased if it is biased for *any* statistic, or just for some set of minimal statistics? - It would be helpful to add the plot showing that the models are already well-calibrated (or at least the reference by Danescu-Mizil and Caruana) in the final version so that the readers are aware of 1) the potential for miscalibration and 2) the fact that these models are in fact well-calibrated.