ECE Colloquium Series – Professor Shimeng Yu
November 30 @ 3:30 pm - 5:00 pm
As part of the *Eleanore Hale Wilson Lecture Series, ECE is proud to present:
Neuro-Inspired Computing with Emerging Non-Volatile Memory (NVM)
Professor Shimeng Yu
Arizona State University
Host: Professor Keshab Parhi
This presentation discusses state-of-the-art, challenges and prospects of the neuro-inspired computing with emerging non-volatile memory (NVM) technologies. First, we discuss the demand for developing neuro-inspired architecture beyond today’s von-Neumann architecture, and we summarize the various approaches to designing the neuromorphic hardware (digital vs. analog, spiking vs. non-spiking, online training vs. offline training). Second, we discuss the desired characteristics of the synaptic devices (e.g. multilevel states, weight tuning linearity, variation/noises), and survey a few representative prototypes reported in the literature that show the “analog” synapse behaviors. Next, we introduce the crossbar array architecture to efficiently implement the weighted sum and weight update operations that are commonly used in the neuro-inspired machine learning algorithms, and review the recent progresses of array-level experimental demonstrations for these key operations. In particular, we discuss “NeuroSim,” a device-circuit-algorithm co-design framework to evaluate the impact of non-ideal device effects on the neuromorphic system performance (i.e. learning accuracy) and trade-offs in the circuit-level performance (i.e. area, latency, energy). Last, we propose to binarize the neural network algorithm to allow very low precision weights and neurons, thereby making it compatible to “digital” synapses (i.e. binary NVM).
Shimeng Yu received the B.S. degree in microelectronics from Peking University, Beijing, China in 2009, and the M.S. degree and Ph.D. degree in electrical engineering from Stanford University, Stanford, CA, USA in 2011 and in 2013, respectively. He is currently an assistant professor of electrical engineering and computer engineering at Arizona State University, Tempe, AZ, USA. His research interests are emerging nano-devices and circuits with a focus on the resistive memories for different applications including machine/deep learning, neuromorphic computing, monolithic 3D integration, hardware security, radiation-hard electronics, etc. He has published >60 journal papers and >100 conference papers with citations >4800 and H-index 29. Among his honors, he is a recipient of the Stanford Graduate Fellowship from 2009 to 2012, the IEEE Electron Devices Society Masters Student Fellowship in 2010, the IEEE Electron Devices Society PhD Student Fellowship in 2012, the DOD-DTRA Young Investigator Award in 2015, the NSF Faculty Early CAREER Award in 2016, and the ASU Fulton Outstanding Assistant Professor in 2017. He did summer internship in IMEC, Belgium in 2011, and IBM TJ Watson Research Center in 2012. He held visiting faculty position in Air Force Research Laboratory in 2016. He served the Technical Program Committee for IEEE International Symposium on Circuits and Systems (ISCAS) 2015-2017, ACM/IEEE Design Automation Conference (DAC) 2017, and IEEE International Electron Devices Meeting (IDEM) 2017.
*Established in 2009, the Eleanore Hale Wilson Fund supports engineering field leaders for travel to Minnesota to share their expertise and discoveries with University of Minnesota graduate students, faculty, and alumni. The fund also supports the receptions held in honor of each speaker.