Bio
Tsung-Te Liu received the B.S. degree in electrical engineering and the M.S. degree in electronics engineering from National Taiwan University, Taipei, Taiwan, in 2002 and 2004, respectively, and the Ph.D. degree in electrical engineering from the University of California at Berkeley, Berkeley, CA, USA, in 2012.
From 2004 to 2005, he was with MediaTek Inc., Hsinchu, Taiwan, where he was involved in circuit and system design for wireless communications. From 2005 to 2012, he was a member of the Berkeley Wireless Research Center (BWRC), University of California at Berkeley. From 2012 to 2014, he was with Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium, where he conducted research on circuit development for advanced CMOS technology. In 2014, he joined the faculty of National Taiwan University, Taiwan, where he is currently an Associate Professor with the Graduate Institute of Electronics Engineering and the Department of Electrical Engineering. He is a recipient of several design and teaching awards. His research interests involve energy-efficient circuit and system designs.
Abstract
In- and Near-Memory Computing (IMC/NMC) architectures have been proposed to substantially enhance the computation efficiencies of machine-learning processing tasks. In this talk, an overview of IMC/NMC will be presented. The concept of IMC/NMC will be first introduced, along with its own design and implementation challenges. After that, the state-of-the-art IMC/NMC solutions and design techniques will be presented. Finally, the design directions of IMC/NMC will be discussed to conclude this talk.