Abstract: Spiking neural networks (SNN), a class of brain-inspired models of computation, are well equipped with spatiotemporal computing power critical for a wide range of applications. Moreover, recent advancements in neuromorphic computing have led to large-scale industrial neuromorphic chips with promising ultra-low energy event-driven data processing capability. Nevertheless, major challenges are yet to be conquered to make spike-based computation a competitive choice for real-world applications. In this talk, first, I will present techniques for tackling major challenges in training complex SNNs by developing biologically plausible learning mechanisms and error backpropagation (BP) operating on top of spiking discontinuities. Second, SNN hardware accelerators, which provide an efficient dedicated computing platform for processing spiking workloads, will be discussed.
Bio: Peng Li received the Ph. D. degree from Carnegie Mellon University in 2003. He is presently a professor of Electrical and Computer Engineering at the University of California, Santa Barbara. His research interests are in integrated circuits and systems, electronic design automation, brain-inspired computing, and applied machine learning. Li’s work has been recognized by an ICCAD Ten Year Retrospective Most Influential Paper Award, four IEEE/ACM Design Automation Conference (DAC) Best Paper Awards, and best paper awards from ICCAD, ICCD, and ASAP, among other distinctions. A Fellow of the IEEE, he served as the Vice President for Technical Activities of IEEE Council on Electronic Design Automation (CEDA).