Intrinsic variation effect in memristive neural network with weight quantization. (10th September 2022)

• Record Type: Journal Article
• Authors: Park, Jinwoo; Song, Min Suk; Youn, Sangwook; Kim, Tae-Hyeon; Kim, Sungjoon; Hong, Kyungho; Kim, Hyungjin
• Is Part Of: Nanotechnology. Volume 33:Number 37(2022)
• Publication Date: 2022-09-10
• Subjects: memristor crossbar array -- memristive neural network -- neuromorphic system -- weight quantization -- off-chip training -- intrinsic variation; Nanotechnology -- Periodicals; Nanotechnology -- Periodicals; Nanotechnology; Publications périodiques; Nanotechnologies; Periodicals; 620.5
• DOI: 10.1088/1361-6528/ac7651
• Languages: English
• ISSNs: 0957-4484

Abstract

Abstract: To analyze the effect of the intrinsic variations of the memristor device on the neuromorphic system, we fabricated 32 × 32 Al2 O3 /TiO x -based memristor crossbar array and implemented 3 bit multilevel conductance as weight quantization by utilizing the switching characteristics to minimize the performance degradation of the neural network. The tuning operation for 8 weight levels was confirmed with a tolerance of ±4 μ A (±40 μ S). The endurance and retention characteristics were also verified, and the random telegraph noise (RTN) characteristics were measured according to the weight range to evaluate the internal stochastic variation effect. Subsequently, a memristive neural network was constructed by off-chip training with differential memristor pairs for the Modified National Institute of Standards and Technology (MNIST) handwritten dataset. The pre-trained weights were quantized, and the classification accuracy was evaluated by applying the intrinsic variations to each quantized weight. The intrinsic variations were applied using the measured weight inaccuracy given by the tuning tolerance, RTN characteristics, and the fault device yield. We believe these results should be considered when the pre-trained weights are transferred to a memristive neural network by off-chip training.