As an important type of deep learning methods, deep generative models provide a suite of flexible tools on revealing the latent structures underlying complex data and performing "top-down" inference to generate samples. In this talk, I will present some recent results on learning with deep generative models, including the basic theory and algorithms, semi-supervised learning with deep generative models, and ZhuSuan--a GPU library to support probabilistic programming and efficient inference.

Dr. Jun Zhu is an associate professor at the Department of Computer Science and Technology in Tsinghua University and an adjunct faculty at Carnegie Mellon University. His research interest lies in probabilistic machine learning. He has published over 100 peer-reviewed papers in the prestigious conferences and journals. He is an associate editor for IEEE Trans. on PAMI and an editorial board member for Artificial Intelligence. He served as area chairs/senior PC's for ICML, NIPS, IJCAI, AAAI and UAI. He was a local co-chair of ICML 2014. He is a recipient of the IEEE Intelligent Systems "AI's 10 to Watch" Award, NSFC Excellent Young Scholar Award, and MIT TR35 China Pioneer Award. His work is supported by the National Youth Top-notch Talents Program.