Congratulations to Dr. Li Sifan, a former PhD student under the supervision of Associate Professor Ang Kah-Wee in the Department of Electrical and Computer Engineering, who has been honored with the Young Researcher Award at the 2023 International Conference on Solid-State Devices and Materials (SSDM).
The conference covers a wide range of subjects related to solid-state devices and materials, including emerging and multidisciplinary research areas. The team presented their research on implementing a neural network hardware accelerator using a two-dimensional (2D) material-based memristive crossbar array at the conference, which was chosen as the recipient of the Young Researcher Award.
The paper entitled “Neural Network Hardware Accelerator using Memristive Crossbar Array based on Wafer-Scale 2D HfSe2” marks a significant milestone as it achieves the first-ever hardware acceleration of neural network computations through the utilization of large memristive crossbar arrays made of 2D hafnium diselenide (HfSe2). In this study, the researchers have successfully developed a method for growing wafer-scale polycrystalline HfSe2 thin films and a metal-assisted van der Waals (vdW) transfer technique. Importantly, this approach can be extended to enable the realization of crossbar arrays using other 2D materials. Furthermore, the crossbar array has been integrated with custom-designed peripheral circuits, enabling the complete hardware implementation of convolutional image processing with narrow current distribution of 0.25%. This research paves the way for the demonstration of large-scale and energy-efficient neuromorphic computing systems for AI applications.