Enhancing LiAlOX synaptic performance by reducing the Schottky barrier height for deep neural network applications. Issue 45 (9th October 2020)

• Record Type: Journal Article
• Authors: Fu, Yaoyao; Dong, Boyi; Su, Wan-Ching; Lin, Chih-Yang; Zhou, Kuan-Ju; Chang, Ting-Chang; Zhuge, Fuwei; Li, Yi; He, Yuhui; Gao, Bin; Miao, Xiang-Shui
• Is Part Of: Nanoscale. Volume 12:Issue 45(2020)
• Page Start: 22970
• Page End: 22977
• Publication Date: 2020-10-09
• Subjects: Nanoscience -- Periodicals; Nanotechnology -- Periodicals; 620.505
• DOI: 10.1039/d0nr04782a
• Languages: English
• ISSNs: 2040-3364

Abstract

Abstract : The synaptic behaviors of LiAlO X devices are optimized by lowering the Schottky barrier height. Abstract : Although good performance has been reported in shallow neural networks, the application of memristor synapses towards realistic deep neural networks has met more stringent requirements on the synapse properties, particularly the high precision and linearity of the synaptic analog weight tuning. In this study, a LiAlO X memristor synapse was fabricated and optimized to address these demands. By delicately tuning the initial conductance states, 120-level continuously adjustable conductance states were obtained and the nonlinearity factor was substantially reduced from 8.96 to 0.83. The significant enhancements were attributed to the reduced Schottky barrier height (SBH) between the filament tip and the electrode, which was estimated from the measured I – V curves. Furthermore, a deep neural network for realistic action recognition task was constructed, and the recognition accuracy was found to be increased from 15.1% to 91.4% on the Weizmann video dataset by adopting the above-described device optimization method.