Generate an Adversary πŸ₯Š

This is a really great introduction to Generative Adversarial Networks (GANs) with great visuals. (and a tutorial if you want to get your hands dirty) πŸ‘©β€πŸ’»πŸ‘¨β€πŸ’»

Why should you care about GANs? They are the driving force behind the crazy creative abilities of neural networks. They are also the building blocks of deepfakes in many cases. πŸ–

What’s the 101 version? GANs pit two neural nets against each other. One creates something, usually an image, that looks real/authentic, like a forger. The other tries to separate real from fake, like an auction house like Christie’s. They go back and forth using the original data and experience from the process to improve and try to beat the other. πŸ₯‡

Src: DataCamp