Ian Goodfellow
Jump to navigation
Jump to search
Ian Goodfellow
An American researcher who invented Generative Adversarial Networks in 2014, reportedly sketching the idea in an evening: train two networks against each other, a generator producing candidate images and a discriminator trying to tell them from real ones, until the generator's output is indistinguishable. GANs made photorealistic synthesis routine and became the technical basis of the first wave of deepfakes. He also co-wrote the standard Deep Learning textbook with Yoshua Bengio and Aaron Courville, and did foundational work on adversarial examples. (See also: Generative AI, Deepfake, Synthetic media, Yoshua Bengio)