Loading Events
Find Events

Event Views Navigation

Upcoming Events

Events List Navigation

September 2017

CESG SEMINAR: “Spiking Neural Networks: Learning Architecture and Hardware Implementation”

September 29 @ 4:10 pm - 5:10 pm
WEB, Room 236-C,
Wisenbaker Engineering Building

Dr. Peng Li, Texas A & M University, Computer Engineering Systems Group Title: “Spiking Neural Networks: Learning Architecture and Hardware Implementation” Abstract: The human brain mediates and produces thoughts, actions, memory, feelings, and many other critical functions with impressive space and energy efficiency. Nevertheless, mimicking the information processing capabilities of the brain in hardware neuromorphic computing systems is a nontrivial task, and entails the development of hardware-friendly learning mechanisms and efficient processor architectures. This talk presents an overview of our recent work on recurrent spiking neural processors based on the model of the liquid state machine (LSM). Spike-dependent learning rules are integrated to support biologically-plausible supervised and unsupervised on-chip learning, leading to self-adaptation. As part of the proposed LSM architecture, runtime reconfiguration is explored to enable firing-activity based power management that employs programmable approximate computing and clock/power gating. Hardware implementation of spiking neural networks (SSNs) on FPGA for pattern and speech recognition and application of SNNs to mobile robot navigation will be also reported. Bio: Dr. Peng Li received the Ph. D. degree in electrical and computer engineering from Carnegie Mellon University. He is a presently a professor of ECE at TAMU and a faculty member of Texas A&M Institute for Neuroscience. His research interests are in integrated circuits and systems, electronic design automation, computational brain modeling, and brain-inspired computing. He has authored and co-authored over 190 publications and edited two books. His work has been recognized by various distinctions including four IEEE/ACM Design Automation Conference Best Paper Awards in 2003, 2008, 2011, and 2016, the ISCAS Honorary Mention Best Paper Award from the Neural Systems and Applications Technical Committee of IEEE Circuits and Systems (CAS) Society in 2016, the IEEE/ACM William J. McCalla ICCAD Best Paper Award in 2012, the National Science Foundation CAREER Award in 2008, two SRC Inventor Recognition Awards, and two MARCO Inventor Recognition Awards. He has consulted for two silicon-valley startup companies and the Intel Corporation, and currently serves the IEEE Council of Electronic Design Automation as its Vice President for Technical Activities. He is a Fellow of the IEEE.

Find out more »
+ iCal Import Listed Events