Memristor-based Neuromorphic Computing Systems

Principal Investigator: Professor Can LI (Assistant Professor from Department of Electrical and Electronic Engineering)

This project is showcased in the third exhibition – Technology for Future.

About the scholar

Professor Can LI

Research interests:
• Artificial intelligence accelerator
• Neuromorphic computing
• Nanoelectronics devices
• Non-volatile memories
• Software-hardware co-optimization


Project information
The evolution of artificial intelligence (AI) and the growing demands from Big Data are hampered by current hardware performance, spurring extensive research into accelerator chips. As silicon transistors reach physical limits, there is an urgent need to explore new computing paradigms based on unconventional devices. Dr Li’s team is developing new brain-inspired computing paradigms using emerging memory devices, aiming to showcase the potential of these neuromorphic computing systems in laboratory settings.
The emerging trend of offloading AI tasks from energy-intensive data centers to edge devices such as smartphones and watches, despite their limited battery and computational capacity. Our solution to this challenge involves computing directly within memory via an advanced microelectronic platform known as memristors, which offer the potential for significant improvements in energy efficiency.
Project images
Integrated Memristive Chip: Microscopic Images and Schematic
Demonstration Unit of Integrated Memristor Chip and Control Board
User Interface of Demonstration Unit: Memristive Neural Network and Optimization
Other projects