Yesterday I was wondering how much I really believe that we are living in a simulation. Today, there is this. You may feel you recognize them, but not one of these faces is a real person, they are all “imagined” using a new frontier in artificial intelligence: a GAN, a neural network where opposing subnetworks compete, creating a feedback loop that escalates into a form of artifical imagination.
This video, just released is impressive and a bit eerie as well.
Tom McKay in Gizmodo described GAN as “a class of algorithm where researchers pair two competing neural networks against each other.”
So one of the two has a generative function, like rendering images or trying to solve a problem, but the other is in adversarial mode, challenging the first’s results.
Two neural networks based on algorithms used in unsupervised machine learning—this in itself, said Hilary Grigonis in Digital Trends, “pushes artificial intelligence to ‘learn’ through trial and error without human intervention, such as separating images of cats and dogs into two groups.”
AI imagination is not a one trick pony, either. After studying old paintings, an AI with this four year old GAN technology created a masterpiece that sold for over $430,000.
Really? Yes, it’s true.