上海科技大学 | English | 地图
首页> 新闻信息> 讲座
Learning Neural Networks with Adaptive Information Flow
Date:2016/12/9     Browse:309
Seminar Topic:  Learning Neural Networks with Adaptive Information Flow

Speaker: Wang Gang
Time: Dec. 9, 3:30 p.m. - 4:30 p.m.
Venue:  Room 1B-106, SIST Building

Abstract: 
Human brains are adept at dealing with the deluge of information they continuously receive, and  adaptively controlling and regulating the information flow to focus on the important inputs and suppress the non-essential ones, for better performance. Inspired by such a capability, we develop three types of networks which support more adaptive information flow in CNN, siamese CNN, and LSTM respectively. Our methods have achieved state-of-the-art performance on CIFAR 100 for image classification, Market-1501 dataset for human re-identification, and NTU RGB-D dataset for action recognition. 

Biography:  
Wang Gang is an Associate Professor with the School of Electrical and Electronic Engineering at Nanyang Technological University (NTU), Singapore. He was an Assistant Professor of the same department from 2010 to 2016. He had a joint appointment at the Advanced Digital Science Center (operated by UIUC) as a research scientist from 2010 to 2014. He received his B.Eng. degree from Harbin Institute of Technology in Electrical Engineering and the PhD degree in Electrical and Computer Engineering, University of Illinois at Urbana-Champaign. His research interests include scene labelling, object recognition, action analysis, and deep learning. He is selected as a regional MIT Technology Review innovator under 35 for Southeast Asia, Australia, New Zealand, and Taiwan. He led a team to achieve top 5 in the ImageNet challenge on scene classification in 2015 and 2016 respectively. He is also a recipient of  Harriett & Robert Perry Fellowship, CS/AI award, best paper awards from PREMIA and top 10 percent paper awards from MMSP.  He is an associate editor of Neurocomputing and has been officially invited as an associate editor for IEEE Transactions on Pattern Analysis and Machine Intelligence.

Seminar 16082