We introduce and motivate generative modeling as a central task for machine learning and provide a critical view of the algorithms which have been proposed for solving this task. We overview how generative modeling can be defined mathematically as trying to make an estimating distribution the same as an unknown ground truth distribution. This can then be quantified in terms of the value of a statistical divergence between the two distributions. We outline the maximum likelihood approach and how it can be interpreted as minimizing KL-divergence. We explore a number of approaches in the maximum likelihood family, while discussing their limitations. Finally, we explore the alternative adversarial approach which involves studying the differences between an estimating distribution and a real data distribution. We discuss how this approach can give rise to new divergences and methods that are necessary to make adversarial learning successful. We also discuss new evaluation metrics which are required by the adversarial approach.
View on arXiv