Abstract: Generative modeling is a key machine learning technique. Recent advances in deep learning and stochastic gradient-based optimization have enabled scalable modeling of high-dimensional complex distributions with impressive applications in image and video generation, text-to-speech synthesis, music generation, and machine translation, among others. In this lecture we will review fundamentals of latent-based and implicit generative models and provide an overview of the key ideas underpinning generative adversarial networks, variational auto-encoders, and autoregressive models.

Speaker bio: Mario Lucic is a senior research scientist at Google Research (Brain team) where he is pursuing fundamental challenges in machine learning and artificial intelligence. He received his Ph.D. in Computer Science from ETH Zurich (2017), a M.Sc. degree (cum laude) in Computer Science from Politecnico di Milano (Italy), and a B.Sc. degree in Computing from University of Zagreb (Croatia). His research on machine learning and artificial intelligence has received awards at several premier conferences, most notably the best paper award at the International Conference on Machine Learning and the best student paper award at the International Conference on Artificial Intelligence and Statistics.