深度学习讲座 | 唐建博士“Learning Representations of Large-scale Networks”

承办单位：网络与信息系统研究所

时间：2017年06月16日（周五）晚上19:30-21:30

地点：理科一号楼1126

主讲人：蒙特利尔大学助理教授唐建博士

 

Learning Representations of Large-scale Networks

Jian Tang, Ph.D.

Assistant Professor

Department of Decision Science, HEC Montreal

Montreal Institute for Learning Algorithms (MILA)

tangjianpku@gmail.com

Abstract

Information networks (e.g., social networks, citation networks,World Wide Web ) are ubiquitous in real world, covering a variety ofapplications. Traditionally, networks are usually represented as adjacencymatrices. However, this type of representation is very sparse andhigh-dimensional, which does not facilitate computation of network analysis andnetwork understanding. In this talk, I will introduce our recent work onlearning low-dimensional representations of large-scale networks including theLINE and LargeVis models. The representations learned by these models are ableto facility a variety of applications such as node classification, nodeclustering, link prediction, recommendation, and network visualization. Thistalk will be a part of my tutorial on network representation learning in thisyear’s KDD.

Speaker’s bio:

Dr. Jian Tang will bejoining the department of decision science, HEC Montreal, as an assistantprofessor starting from this fall. He will also be a faculty member of MontrealInstitute for Learning Algorithms (MILA), which is the deep learning group leadby one of the deep learning pioneers Yoshua Bengio. His research interests aredeep learning, reinforcement learning, statistical topic modelling with variousapplications. He was a research fellow in University of Michigan and CarnegieMellon University. He received his Ph.D degree from Peking University and was anassociate researcher in Microsoft Research Asia. He received the best paperaward of ICML’14 and nominated for the best paper of WWW’16. He is a PC memberof many prestigious conferences such as IJCAI, AAAI, ACL, EMNLP, WWW, WSDM, andKDD.

演讲人简介：

唐建博士今年秋季将加入蒙特利尔大学商学院担任助理教授，同时也是深度学习奠基人之一Yoshua Bengio教授领导的深度学习小组MILA的教师成员。 他的研究兴趣包括深度学习、强化学习、统计主题模型以及这些方法在不同领域的应用。他于2014年北京大学信息科学技术学院获得博士学位。曾经是密歇根大学和卡内基梅隆大学的联合培养博士后以及微软亚洲研究院副研究员。他于2014年获得机器学习顶级会议ICML的最佳论文以及2016年获得数据挖掘顶级会议WWW最佳论文提名。他是人工智能和数据挖掘领域多个顶级会议的程序委员如IJCAI, AAAI, ACL, EMNLP, WWW, WSDM, and KDD.

6.16晚理科一号楼1126，敬请期待！！！

唐建博士6月8日讲座回顾：

6月8日，唐建博士为大家带来“An Introduction to Deep Learning & How to Do Research in Machine Learning”主题讲座，现场很多学生慕名而来，100多人的教室座无虚席。唐建博士深入浅出地介绍了深度学习领域的相关技术，并现场解答学生的众多问题。讲座的最后，为大家传授如何做机器学习领域的科学研究。

附6月8日讲座现场图