This new version uses an additional model that allows for a 1024×1024 output and adds more details like skin texture, eyebrows and eyelashes. There are still several technical issues that I need to find a solution for, like for example the repetitive patterns that appear in low-entropy areas.
Also check out the great talk about the topic.
Generative adversarial network
A generative adversarial network (GAN) is a class of machine learning frameworks designed by Ian Goodfellow and his colleagues in 2014. Two neural networks contest with each other in a game (in the sense of game theory, often but not always in the form of a zero-sum game). Given a training set, this technique learns to generate new data with the same statistics as the training set. For example, a GAN trained on photographs can generate new photographs that look at least superficially authentic to human observers, having many realistic characteristics. Though originally proposed as a form of generative model for unsupervised learning, GANs have also proven useful for semi-supervised learning, fully supervised learning, and reinforcement learning.
Definition from Wikipedia – Generative adversarial network