EPSCoR Research Fellows: NSF: Memristor Based Computing-in-Memory for Neuromorphic System: Design, Optimization, and Fabrication

Generative AI, such as ChatGPT, along with AI image and video generation platforms, have recently taken the world by storm. However, recent studies have revealed that running AI engines consumes a staggering amount of energy. Neuromorphic systems, which utilize memristors and crossbar arrays, leverage Computing-in-Memory (CIM) technology for AI computation, demonstrating significant improvements in energy efficiency. The goal of this project is to investigate memristor-based CIM for neuromorphic systems from the perspectives of design, optimization, and fabrication. This project offers a unique opportunity for the principal investigator (PI) from the Department of Electrical and Computer Engineering (ECE) at the University of South Alabama (USA) to establish a long-term collaboration with the School of ECE at the Georgia Institute of Technology (Georgia Tech), thereby enhancing the PI’s research capabilities. This collaboration will not only broaden the PI’s research scope but also greatly benefit his career trajectory, ultimately contributing significantly to his home institution and jurisdiction during and beyond the two-year award period.

This fellowship provides the PI with an excellent opportunity to explore novel AI computing systems from a hardware perspective, thus steering his research toward transformative new directions. It will also significantly contribute to the economic and technological development of this EPSCoR-eligible jurisdiction. The PI will share the knowledge and resources obtained from Georgia Tech with his colleagues and the broader community, thereby raising the overall research and educational capacity and competitiveness of his institution and jurisdiction. Based on the knowledge acquired at Georgia Tech, the PI will also organize workshops and seminars at his home institution and jurisdiction to foster broad collaborations and maximize the benefits of this fellowship. The enhanced knowledge gained through this fellowship will serve as a crucial link, connecting scientific research with practical applications in AI hardware design, and facilitating the development of future novel computing systems.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2611357
Principal Investigator: Jinhui Wang
Funds Obligated: $296,037
State: AL