A Probe Towards Understanding GAN and VAE Models
This is an incremental study providing personal interpretations and a new model for understanding generative models, relevant to researchers in machine learning.
The paper compares GAN and VAE models prior to 2017, summarizing experiments on fidelity and mode collapse, and proposes a new model based on a hypothesis to explain their differences, tested on MNIST and CelebA datasets.
This project report compares some known GAN and VAE models proposed prior to 2017. There has been significant progress after we finished this report. We upload this report as an introduction to generative models and provide some personal interpretations supported by empirical evidence. Both generative adversarial network models and variational autoencoders have been widely used to approximate probability distributions of data sets. Although they both use parametrized distributions to approximate the underlying data distribution, whose exact inference is intractable, their behaviors are very different. We summarize our experiment results that compare these two categories of models in terms of fidelity and mode collapse. We provide a hypothesis to explain their different behaviors and propose a new model based on this hypothesis. We further tested our proposed model on MNIST dataset and CelebA dataset.