Artificial Neurons Using Ag−In−Zn−S/Sericin Peptide‐Based Threshold Switching Memristors for Spiking Neural Networks

Nan He,Jie Yan,Zhining Zhang,Haiming Qin,Ertao Hu,Xinpeng Wang,Hao Zhang,Pu Chen,Feng Xu,Yang Sheng,Lei Zhang,Yi Tong

DOI: https://doi.org/10.1002/aelm.202300467

IF: 6.2

2023-09-22

Advanced Electronic Materials

Abstract:The studied Ag/Ag−In−Zn−S/silk sericin/W device demonstrates excellent electrical properties and is capable of emulating leaky integrate‐and‐fire (LIF) neuron functionality and strength‐modulated spike frequency characteristics without the need for additional auxiliary circuits. A three‐layer spiking neural networks demonstrated by simulation realizes a high classification accuracy of 91.44% for MNIST handwritten digit based on the experimentally measured LIF data. Memristive devices with threshold switching characteristics can be effectively utilized to mimic biological neurons acting as one of the key building blocks for constructing advanced hardware neural networks. In this work, the emulation of leaky integrate‐and‐fire memristive neuron is realized in one single cell with Ag/Ag−In−Zn−S/silk sericin/W architecture without the need for additional auxiliary circuits. The studied devices demonstrate excellent electrical properties, such as stably repeatable threshold switching, concentratedly low threshold voltage (≈0.4 V), and relatively small device‐to‐device variation. In addition, multiple neural features, such as leaky integrate‐and‐fire neuron functionality and strength‐modulated spike frequency characteristic, have been successfully emulated owing to the forming‐free volatile threshold switching effect. The stable volatile threshold switching behaviors and regular firing event may be attributed to the controllable metallic Ag filamentary mechanism. Furthermore, a solid accuracy of 91.44% of the pattern recognition of Modified National Institute of Standards and Technology (MNIST) data is obtained via a trained spiking neural network (SNN) based on the leaky integrate‐and‐fire behavior of sericin‐based device. These achievements shed light on the fact that employing sericin biomaterials has great application potential in advanced neuromorphic computation.

materials science, multidisciplinary,physics, applied,nanoscience & nanotechnology

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to develop a new type of artificial neuron device to overcome the limitations of large area and low efficiency of traditional complementary metal - oxide - semiconductor (CMOS) neurons when implementing large - scale neural networks. Specifically, the authors designed a threshold - switching memristor based on silver - indium - zinc - sulfur/sericin (Ag − In − Zn − S/Sericin). This device can simulate the function of a leaky - integrate - and - fire (LIF) neuron in a single unit without additional auxiliary circuits. The research shows that these devices have stable threshold - switching characteristics, a low operating voltage (about 0.4V), good device - to - device consistency, and diverse neuron characteristics, such as LIF function and intensity - modulated pulse - frequency characteristics. In addition, through a spiking neural network (SNN) trained based on LIF behavior, a recognition accuracy of 91.44% for the MNIST handwritten digit dataset was achieved, indicating the great potential of using sericin materials in the hardware implementation of advanced neuromorphic computing systems.