Weiqiang Liu
Speaker at our Cybersecurities of the Future event
Weiqiang Liu is currently a full Professor and the Vice Dean of College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics (NUAA), Nanjing, China. He received the B.Sc. degree in Information Engineering from NUAA and the Ph.D. degree in Electronic Engineering from Queen's University Belfast (QUB), Belfast, United Kingdom, in 2006 and 2012, respectively. He has served as an Associate Editor of IEEE Transactions on Computers (TC), IEEE Transactions on Circuits and Systems I: Regular Papers (TCAS-I), IEEE Transactions on Emerging Topics in Computing (TETC), a Steering Committee Member of IEEE Transactions on VLSI Systems, the Guest Editors of Proceedings of the IEEE. He is the Program Committee Co-Chair of IEEE ARITH 2020. He has been a technical program committee member for a number of international conferences. His research interests include approximate computing, emerging technologies in computing systems, computer arithmetic, and hardware security. He has published one research book by Artech House and over 130 leading journal and conference papers. One of his papers was selected as the Highlight Paper of IEEE TCAS-I 2021 Janurary Issue and the Feature Paper of IEEE TC in the 2017 December issue. He has been awarded the prestigious Excellent Young Scholar Award by National Natural Science Foundation of China (NSFC) in 2020. He is a Senior Member of the IEEE, CIE and CCF.
Abstract
Approximate computing has been proposed for highly energyefficient systems targeting emerging error tolerant applications such as machine learning and digial signal processing. Approximate computing consists of approximately (inexactly) processing data to save power and achieve high performance, while results remain at an acceptable level for subsequent use. In this talk, I will first introduce approximate computing and review the current approximate desgins which mainly focus on computer arithmetic. Then, the algorithm and approximate circuit co-design method for deep neural networks is studied. Finally, the security of approximate computing itself will also be discussed.
Abstract
Approximate computing has been proposed for highly energyefficient systems targeting emerging error tolerant applications such as machine learning and digial signal processing. Approximate computing consists of approximately (inexactly) processing data to save power and achieve high performance, while results remain at an acceptable level for subsequent use. In this talk, I will first introduce approximate computing and review the current approximate desgins which mainly focus on computer arithmetic. Then, the algorithm and approximate circuit co-design method for deep neural networks is studied. Finally, the security of approximate computing itself will also be discussed.