GAN and VAE from an Optimal Transport Point of View

Abstract

This short article revisits some of the ideas introduced in arXiv:1701.07875 and arXiv:1705.07642 in a simple setup. This sheds some lights on the connexions between Variational Autoencoders (VAE), Generative Adversarial Networks (GAN) and Minimum Kantorovitch Estimators (MKE).

Cited by in corpus (7)

• On Wasserstein Reinforcement Learning and the Fokker-Planck equation
• Conditional deep surrogate models for stochastic, high-dimensional, and multi-fidelity systems
• A gradual, semi-discrete approach to generative network training via explicit Wasserstein minimization
• k-GANs: Ensemble of Generative Models with Semi-Discrete Optimal Transport
• Progressive VAE Training on Highly Sparse and Imbalanced Data
• A Chain Graph Interpretation of Real-World Neural Networks
• Loosely Coupled Federated Learning Over Generative Models