Sicong Liu

Email: scliu007@gmail.com

Address: No.2 South Taibai Road, Yanta Zone, Xi'an City, Shaanxi Province, PRC

Personal Statement

I'm a third-year Ph.D. student from Xidian University, under the supervision of Prof. Junzhao Du. My research interests include intelligent awareness on mobile and embedded devices by using the machine learning, optimization, reinforcement learning and deep learning techniques. Currently I’m focusing on the on-demand deep model compression for different requirements of recognition tasks and diverse resource limitations of mobile devices.

Education Background

Sep. 2017 - Sep. 2018, Joint-supervised PhD student, Department of Electrical and Computer Engineering, Rice University, visiting PhD student, hosted by Prof. Lin Zhong

Jan. 2016 - now, school of computer science and technology, Xidian University, PhD student (rank first in the major), hosted by Prof. Junzhao Du

Sep.2013-Dec. 2015, school of computer science and technology, Xidian University, Master Degree (rank first in the major), hosted by Prof. Junzhao Du

Sep. 2009- Jul. 2013, School of Software engineering, Xidian University, Bachelor Degree (top 15% in the major)

Activities and Awards

Given a talk "Embedded and Mobile Deep Learning Design" in Google brain & Tensorflow Lite team and Alibaba, 2018 Aug.

Networking Networking Women (N2 Women) Fellowship, 2018 (1)

Graduate Student Innovation Fund from Xidian University, 2018

State Schoolarship Fund from China Schoolarship Council, 2017

Assistant the organization of the ACM TURC 2017 conference and won the “Student Volunteer Award"

China national scholarship in 2017, 2015, 2014 (1%)

Mobisys 2016 student travel grant, have a Poster Display in this conference and assist the organization of Asian Students Symposium on Emerging Technologies, 2016

Huawei student scholarship with the title of "technical master", 2015

Won the "silver" award in Tencent QRobot development and application contest with the "Magic Tone" project, 2014

Publications

Sicong Liu, Yingyan Lin, Zimu Zhou, Kaiming Nan, Hui Liu, Junzhao Du, On-demand deep model compression for mobile devices a usage-driven model selection framework, Proceedings of the 16th Annual International Conference on MobiSys 2018.

Sicong Liu, Zimu Zhou, Junzhao Du, LongFei Shangguan, Jun Han, UbiEar: bringing location-independent sound awareness to the hearing impaired with smartphones, IMWUT Journal, Ubicomp 2017.

Sicong Liu, Hui Liu, Junzhao Du, Hui Liu, Rui Li, Xue yang, Keiwei Sha. Energy-efficient algorithm to construct the information potential field in WSNs, IEEE Sensor Journal 2017.

Sicong Liu, Junzhao Du. Poster: MobiEar-building an environment-independent acoustic sensing platform for the deaf using deep learning, Proceedings of the 14th Annual International Conference on MobiSys 2016. ACM, 2016: 50-50.

Sicong Liu, Junzhao Du, Xue Yang, Rui Li, Hui Liu, Kewei Sha, CrowdBlueNet: maximizing crowd data collection using multi-hop Bluetooth networks, WASA 2016.

Xue Yang, Junzhao Du, Sicong Liu, Rui Li, Hui Liu. Air pollution source estimation profiling via mobile sensor networks, CITS 2016

Hui Liu, Rui Li, Sicong Liu, Shibian Tian, Juzhao Du. SmartCare: energy-efficient long-term physical activity tracking using smartphones. Tsinghua Science and Technology journal, TST Journal 2015.

Junzhao Du, Sicong Liu, Chi Xu, Kai Wang, Hui Liu, Kewei Sha. Lightweight construction of the information potential field in wireless sensor networks. Computer Communication and Networks, ICCCN 2014.

Research Interest

Deep learning based mobile applications

Application uniquely-tied deep learning optimization

Automatic and adaptive deep model compression for mobile and embedded devices