Physics Colloquium: Prototyping Emerging Devices and Networks for Neuro-Inspired Computing

Prof. Gina Adam, George Washington University

Abstract: Artificial intelligence systems are expected to consume massive amounts of computing resources in the coming decades at significant financial and environmental costs. New hardware alternatives are necessary to keep up with the increasing demand in complexity and energy efficiency in this sector. Emerging analog memory devices, like resistive switching technology (ReRAM or memristor) have shown potential for the compact and efficient implementation of artificial synapses for neuro-inspired computing. However, prototyping a neural network system with these devices is not an easy task given their issues related to manufacturability, reliability and yield. This talk describes the steps in design, fabrication and CMOS integration of oxide-based memristor switches and their applicability for neural network systems. The nanofabrication and characterization approaches used to optimize the sub-stoichiometric TiO2-x active material and the device design for improved performance and yield at the wafer scale will be presented. Results on the performance of these devices at the population level will be shown. To support these efforts, a modular mixed-signal prototyping platform was collaboratively designed for benchmarking memresistive neural networks of up to 20,000 two-terminal devices. Parallel neural network inference results on a multitude of pre-trained ternary weight solutions for classification will be discussed. Moreover, a complementary simulation model will be shown to accurately predict the experimental hardware performance, showcasing the effectiveness of the proposed system for device-algorithm co-design.

Gina Adam is an assistant professor with the Electrical and Computer Engineering department at the George Washington University. Her group works on the development of emerging non-volatile memory devices and novel hardware foundations that will enable new ways of neuro-inspired computing. She received her Ph.D. in electrical and computer engineering from the University of California Santa Barbara in 2015 and was a research scientist at the Romanian National Institute for Research and Development in Microtechnologies and a visiting scholar at École Polytechnique Fédérale de Lausanne before joining GWU. She was the recipient of an International Fulbright Science and Technology award in 2010, a Mirzayan fellowship at the National Academy of Engineering in 2012, a H2020 Marie Sklodowska-Curie grant from the European Commission in 2016, a NSF CRII award in 2020, and AFOSR YIP and NSF CAREER awards in 2023.