An artificial neural network chip based on two-dimensional semiconductor

Shunli Ma; Tianxiang Wu; Xinyu Chen; Yin Wang; Hongwei Tang; Yuting Yao; Yan Wang; Ziyang Zhu; Jianan Deng; Jing Wan; Ye Lu; Zhengzong Sun; Zihan Xu; Antoine Riaud; Chenjian Wu; David Wei Zhang; Yang Chai; Peng Zhou; Junyan Ren; Wenzhong Bao

doi:10.1016/j.scib.2021.10.005

An artificial neural network chip based on two-dimensional semiconductor

Sci Bull (Beijing). 2022 Feb 15;67(3):270-277. doi: 10.1016/j.scib.2021.10.005. Epub 2021 Oct 5.

Authors

Shunli Ma¹, Tianxiang Wu¹, Xinyu Chen¹, Yin Wang¹, Hongwei Tang¹, Yuting Yao¹, Yan Wang¹, Ziyang Zhu², Jianan Deng², Jing Wan², Ye Lu², Zhengzong Sun¹, Zihan Xu³, Antoine Riaud¹, Chenjian Wu⁴, David Wei Zhang¹, Yang Chai⁵, Peng Zhou⁶, Junyan Ren⁷, Wenzhong Bao⁸

Affiliations

¹ State Key Laboratory of ASIC and System, School of Microelectronics, Fudan University, Shanghai 200433, China.
² State Key Laboratory of ASIC and System, School of Information Science and Technology, Fudan University, Shanghai 200433, China.
³ Shenzhen Sixcarbon Technology, Shenzhen 518106, China.
⁴ School of Electronic and Information Engineering, Soochow University, Suzhou 215006, China.
⁵ Department of Applied Physics, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China.
⁶ State Key Laboratory of ASIC and System, School of Microelectronics, Fudan University, Shanghai 200433, China. Electronic address: pengzhou@fudan.edu.cn.
⁷ State Key Laboratory of ASIC and System, School of Microelectronics, Fudan University, Shanghai 200433, China. Electronic address: junyanren@fudan.edu.cn.
⁸ State Key Laboratory of ASIC and System, School of Microelectronics, Fudan University, Shanghai 200433, China. Electronic address: baowz@fudan.edu.cn.

PMID: 36546076
DOI: 10.1016/j.scib.2021.10.005

Abstract

Recently, research on two-dimensional (2D) semiconductors has begun to translate from the fundamental investigation into rudimentary functional circuits. In this work, we unveil the first functional MoS₂ artificial neural network (ANN) chip, including multiply-and-accumulate (MAC), memory and activation function circuits. Such MoS₂ ANN chip is realized through fabricating 818 field-effect transistors (FETs) on a wafer-scale and high-homogeneity MoS₂ film, with a gate-last process to realize top gate structured FETs. A 62-level simulation program with integrated circuit emphasis (SPICE) model is utilized to design and optimize our analog ANN circuits. To demonstrate a practical application, a tactile digit sensing recognition was demonstrated based on our ANN circuits. After training, the digit recognition rate exceeds 97%. Our work not only demonstrates the protentional of 2D semiconductors in wafer-scale integrated circuits, but also paves the way for its future application in AI computation.

Keywords: Artificial neural network (ANN); Circuits; MoS(2); Multiply-and-accumulate (MAC); Two-dimensional (2D) FETs.

MeSH terms

Children's Health Insurance Program*
Computer Simulation
Molybdenum*
Neural Networks, Computer
Semiconductors

Substances

Molybdenum