本次活动教师发展中心联合信息地学研究中心特别邀请哥伦比亚大学博士后刘小洋，与我校师生分享其在张量理论、大数据分析与深度学习方面的研究。具体安排如下：

一、主  题：Tensors: Big Data & Deep Learning

二、主讲人：哥伦比亚大学 博士后 刘小洋

三、时  间：2017年6月6日（周二）9:30-11:00

四、地  点：清水河校区经管楼宾诺咖啡

五、主持人：通信与信息工程学院 钱峰 副教授

六、交流内容

Driving by a rapidly growing number of sensor devices and sensing systems of rapidly growing resolution, the big data era is swamping areas including data analysis, deep learning, signal processing, scientific computing, and cloud computing. Higher-order tensor modeling together with efficient tensor-based algorithms enables such a fundamental paradigm shift. Data analysis and deep learning architectures upon tensor models have great flexibility in the choice of constraints that match data properties, thus extract more meaningful latent components than matrix-based methods.

Two novel tensor models are proposed, with successfully applications such as sensory and seismic data processing, WiFi fingerprint-based indoor localization, drone-based wireless relay, MRI imaging, video compression and denoising, and image clustering. Promising applications are in deep learning, with three examples: two-dimensional dictionary learning, face recognition, and neural networks for real-time localization of smartphones.

七、主讲人简介

Xiao-Yang Liu received his BE. degree from the School of Computer Science and Technology, Huazhong University of Sci. & Tech., China, in 2010  and his PhD. degree from the Department of Computer Science and Engineering, Shanghai Jiao Tong University in 2016.  Now, he is working as a postdoctoral research fellow in Department of Electrical Engineering at the Columbia University. His research interests include tensor theory, deep learning, big data analysis, cloud computing, and networking.

Dr. Liu has a total of over 400 citations according to Google Scholar Citations. His TPDS 2015 paper is an ESI highly-cited paper. He published papers in top journals such as IEEE TMC, TDSC, TPDS, TITS, TVT, Elsevier Neurocomputing, etc.; and in top conferences such as INFOCOM, ICDCS, MobiHoc, and ICME.

八、主办单位：人力资源部教师发展中心

承办单位：信息地学研究中心、通信与信息工程学院