Topic 10: Generative Adversarial Neural Networks (GANs)
This chapter introduces GANs, application of deep generative models and challenges in optimization.
-
Chapter 10.01: Introduction to Generative Models
In this section, we introduce the generative model, a powerful family of machine learning.
-
Chapter 10.02: Probabilistic graphical models
In this section, we explain probabilistic graphical models, latent variables, and directed graphical models.
-
Chapter 10.03: Tuning with Variation Autoencoders
Instead of mapping the input into a fixed vector, we want to map it into a distribution.
-
Chapter 10.04: Generative Adversarial Networks (GANs)
Generative adversarial networks (GANs) are an exciting recent innovation in machine learning. GANs are generative models: they create new data instances that resemble your training data. For example, GANs can create images that look like photographs of human faces, even though the faces don’t belong to any real person.
-
Chapter 10.05: GAN variants
We explain non-saturating loss and conditional GANs in this part.
-
Chapter 10.06: Challenges for GAN Optimization
We explain challenges of GANs model such as no convergence to fix point as well as problems of adversarial setting.