T4:Deep Generative Models, Algorithms and Platforms
讲者: 朱军(清华)
Abstract: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 tutorial, I will present some results on learning with deep generative models, including the basic theory and algorithms of deep generative models (e.g., VAE and GAN), supervised and semi-supervised learning with deep generative models (e.g., max-margin learning and triple-GAN), a new conditional moment-matching criterion to estimate the parameters, and ZhuSuan--a GPU library to support probabilistic programming and efficient inference. (讲义内容为英文,报告语言为中文。)
个人简介:朱军,清华大学计算机系长聘副教授、卡内基梅隆大学兼职教授、智能技术与系统国家重点实验室副主任。2001到2009年获清华大学计算机学士和博士学位,之后在卡内基梅隆大学做博士后,2011年回清华任教。主要从事机器学习基础理论、高效算法及相关应用研究,在国际重要期刊与会议发表学术论文近百篇。受邀担任人工智能顶级杂志IEEE TPAMI和AI的编委、《自动化学报》编委,担任机器学习国际大会ICML2014地区联合主席, ICML (2014-2017)、NIPS (2013, 2015)、UAI (2014-2017)、IJCAI(2015,2017)、AAAI(2016, 2017)等国际会议的领域主席,中国计算机学会(CCF)学术工委主任助理。获CCF青年科学家奖、国家优青基金、北京市优秀青年人才奖、清华大学优秀班主任一等奖等,入选国家“万人计划”青年拔尖人才、IEEE Intelligent Systems杂志评选的“AI’s 10 to Watch”、及清华大学221基础研究人才计划。