|
|||||
|
|
A Reliable All-2D Materials Artificial Synapse for High Energy-Efficient Neuromorphic Computing |
|
| Authors: | Jian Tang Congli He Jianshi Tang Kun Yue Qingtian Zhang Yizhou Liu Qinqin Wang Shuopei Wang Na Li Cheng Shen Yanchong Zhao Jieying Liu Jiahao Yuan Zheng Wei Jiawei Li Kenji Watanabe Takashi Taniguchi Dashan Shang Shouguo Wang Wei Yang Rong Yang Dongxia Shi Guangyu Zhang |
| Affiliation: | 1. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China;2. Institute of Advanced Materials, Beijing Normal University, Beijing, 100875 China;3. Institute of Microelectronics, Beijing National Research Center for Information Science and Technology (BNRist), Tsinghua University, Beijing, 100084 China
Beijing Innovation Center for Future Chips (ICFC), Tsinghua University, Beijing, 100084 China;4. Ming Hsieh Department of Electrical Engineering, University of Southern California, Los Angeles, CA, 90089 USA;5. Beijing Innovation Center for Future Chips (ICFC), Tsinghua University, Beijing, 100084 China;6. RIKEN Center for Emergent Matter Science (CEMS), Wako, 351-0198 Japan;7. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China School of Physical Sciences, University of Chinese Academy of Sciences, Beijing, 100190 China;8. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China Songshan Lake Materials Laboratory, Dongguan, Guangdong, 523808 China;9. Research Center for Functional Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba, 305-0044 Japan;10. International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, 305-0044 Japan;11. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China School of Physical Sciences, University of Chinese Academy of Sciences, Beijing, 100190 China Songshan Lake Materials Laboratory, Dongguan, Guangdong, 523808 China The Key Laboratory of Microelectronics Device & Integrated Technology, Institute of Microelectronics, Chinese Academy of Sciences, Beijing, 100029 China;12. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China School of Physical Sciences, University of Chinese Academy of Sciences, Beijing, 100190 China Songshan Lake Materials Laboratory, Dongguan, Guangdong, 523808 China Beijing Key Laboratory for Nanomaterials and Nanodevices, Beijing, 100190 China;13. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China Songshan Lake Materials Laboratory, Dongguan, Guangdong, 523808 China Beijing Key Laboratory for Nanomaterials and Nanodevices, Beijing, 100190 China;14. Beijing National Laboratory for Condensed Matter Physics, Key Laboratory for Nanoscale Physics and Devices, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190 China School of Physical Sciences, University of Chinese Academy of Sciences, Beijing, 100190 China Beijing Key Laboratory for Nanomaterials and Nanodevices, Beijing, 100190 China |
| Abstract: | High-performance artificial synaptic devices are indispensable for developing neuromorphic computing systems with high energy efficiency. However, the reliability and variability issues of existing devices such as nonlinear and asymmetric weight update are the major hurdles in their practical applications for energy-efficient neuromorphic computing. Here, a two-terminal floating-gate memory (2TFGM) based artificial synapse built from all-2D van der Waals materials is reported. The 2TFGM synaptic device exhibits excellent linear and symmetric weight update characteristics with high reliability and tunability. In particular, the high linearity and symmetric synaptic weight realized by simple programming with identical pulses can eliminate the additional latency and power consumption caused by the peripheral circuit design and achieve an ultralow energy consumption for the synapses in the neural network implementation. A large number of states up to ≈3000, high switching speed of 40 ns and low energy consumption of 18 fJ for a single pulse have been demonstrated experimentally. A high classification accuracy up to 97.7% (close to the software baseline of 98%) has been achieved in the Modified National Institute of Standards and Technology (MNIST) simulations based on the experimental data. These results demonstrate the potential of all-2D 2TFGM for high-speed and low-power neuromorphic computing. |
| Keywords: | 2D materials artificial synapse linear weight update MoS 2 |
