Bias And Generalization In Deep Generative Models: An Empirical Study

Authors:
Shengjia Zhao Stanford University
Hongyu Ren Stanford University
Arianna Yuan Stanford University
Jiaming Song Stanford University
Noah Goodman Stanford University
Stefano Ermon Stanford

Introduction:

In high dimensional settings, density estimation algorithms rely crucially on their inductive bias.In this paper the authors propose a framework to systematically investigate bias and generalization in deep generative models of images by probing the learning algorithm with carefully designed training datasets.

Abstract:

In high dimensional settings, density estimation algorithms rely crucially on their inductive bias. Despite recent empirical success, the inductive bias of deep generative models is not well understood. In this paper we propose a framework to systematically investigate bias and generalization in deep generative models of images by probing the learning algorithm with carefully designed training datasets. By measuring properties of the learned distribution, we are able to find interesting patterns of generalization. We verify that these patterns are consistent across datasets, common models and architectures.

You may want to know: