The pursuit of better memory and computing systems: from ions to chips

Wei Lu has been named the James R. Mellor Professor of Engineering at the University of Michigan. A reception will immediately follow the talk.

The surge of data-intensive applications, particularly in Artificial Intelligence (AI), along with challenges in transistor scaling, has prompted the semiconductor industry to explore new devices and computing architectures. This presentation provides an overview of our 19-year journey at UM in memory device and computing system research.
Starting from a single device, we revealed the ability to manipulate ions in semiconductors and insulators using voltage, resulting in distinct resistance states for information storage. These devices can serve as high-density memory units and can be directly integrated with existing silicon chips. Furthermore, the underlying physics of the device operation can be directly used for computation, facilitating the efficient execution of both deep neural networks (DNNs) and bio-inspired neural networks. These memory and computing chips have now made the transition from research to commercialization. New research directions that aim for further improving efficiency, latency and reliability will also be discussed.

Bio

Wei Lu is a Professor in the Electrical Engineering and Computer Science department at the University of Michigan. He received B.S. in physics from Tsinghua University, Beijing, China, in 1996, and Ph.D. in physics from Rice University, Houston, TX in 2003. From 2003 to 2005, he was a postdoctoral research fellow at Harvard University, Cambridge, MA. He joined UM faculty in 2005. His research interests include memory devices, in-memory computing and neuromorphic computing systems, and low-dimensional devices. To date Prof. Lu has published over 200 journal and conference articles with 40,000 citations and h-index of 90. He is an IEEE Fellow, and co-founder of Crossbar Inc, and MemryX Inc.

SPONSORED BY
Electrical and Computer Engineering