From the course: Introduction to Generative Adversarial Networks (GANs)

Origin: Where did they come from?

From the course: Introduction to Generative Adversarial Networks (GANs)

Origin: Where did they come from?

“

- [Instructor] Where exactly did Generative Adversarial Networks come from? Where did that idea come from? Well, there's quite a deep history of research around these concepts, and a lot of that research has contributed to where we are with GANs today to where we can see their generative power. A lot of thoughts initially came from different philosophy. So people were talking about these concepts of having machines being able to work together and against each other to be able to generate new images or generate new ideas, but when we talk technically about where this concept came from, at least for myself, the first paper where I heard about it was from this research paper called Generative Adversarial Nets. This research really kicked off and showed what was possible in terms of image generation. I have a few examples of the images from the paper, and they show how handwritten digits, the celebrity faces, even small little images could be generated using some of the methods that they explain. So some of those core concepts, we'll go through in this course and we'll talk about how you can train your own basic model using one of these example data sets.

Contents