Generative Models 101
Although in typical applications of machine learning data is seen as in input to a model, generative models output (or generate) it. These models have been extensively used to test our current capabilities and, in some cases, they help guiding our intuitions towards better architectures. In my talk, I'll give a brief overview of generative models, and give examples on text, images, and beyond.
Oriol Vinyals is a Research Scientist at DeepMind. Previously he was a member of the Google Brain team. He holds a Ph.D. in EECS from University of California, Berkeley, and a Masters degree from University of California, San Diego.