Garrett S. Rose is a Professor and Department Head in the Min H. Kao Department of Electrical Engineering and Computer Science at the University of Tennessee, Knoxville where his research is focused in the areas of nanoelectronic circuit design, neuromorphic computing and hardware security. Prior to joining UT, from June 2011 to July 2014, he was with the Air Force Research Laboratory, Information Directorate, Rome, NY. He received the B.S. degree in computer engineering from Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, in 2001 and the M.S. and Ph.D. degrees in electrical engineering from the University of Virginia, Charlottesville, in 2003 and 2006, respectively. His Ph.D. dissertation was on the topic of circuit design methodologies for molecular electronic circuits and computing architectures.
Presentation Title:
CMOS-Memristive Neuromorphic Systems: Pathways for Next Generation, Energy-Efficient Architectures
Presentation Abstract:
Neuromorphic computing, which operates using spiking information, has emerged as an energy-efficient option for brain-inspired computing, especially well-suited for devices on the edge. As is the case with many emerging architectures, neuromorphic systems may also find further efficiencies through implementations built around memristive devices and systems. Such memristive implementations are particularly attractive for their improved memory density and behavior that in many ways mimics that of biological synapses. However, memristor-based computing, neuromorphic included, still requires conventional CMOS transistor-based electronics in order realize complete circuits and systems. This talk will explore some of the CMOS-memristive neuromorphic system designs that have been considered over the years. We will also discuss paths forward for such technologies, including how hybrid CMOS-memristive technologies, perhaps even utilizing other exotic device technologies, may be able to provide much needed energy-efficient solutions for future AI systems.