{"title":"Deep spiking neural networks with integrate and fire neuron using steep switching device","authors":"Sung Yun Woo , Sangyeon Pak , Sung-Tae Lee","doi":"10.1016/j.sse.2024.108860","DOIUrl":null,"url":null,"abstract":"<div><p>Deep learning has shown impressive capabilities in tasks like speech recognition and image classification. However, modern deep neural networks often demand a significant number of weights and extensive computational resources, creating efficiency challenges for applications on edge devices. To address these issues, researchers have introduced deep spiking neural networks (DSNNs) that leverage specialized hardware for synapses and neurons. DSNNs offer a potential solution by improving efficiency in edge-device applications. In this paper, the hardware based DSNN with integrate and fire neuron using steep switching device was investigated. We propose integrate and fire neuron using steep switching device to implement rate coding as input encoding method. Because the steep switching device has double-gate, the threshold voltage of the neuron circuits can be adaptively controlled, which changes the rates of input pulse. Hence, the adjustment of the threshold of neuron can be employed to mitigate the accuracy deterioration resulting from the transformation from deep neural networks (DNNs) to DSNNs. In addition, the off-current of proposed integrate and fire neuron circuit decreases significantly as the steep switching device has steep subthreshold swing. A system simulation of a hardware based DSNN shows that the adjustable threshold of the neuron circuit can achieve a high inference accuracy of 98.36 % which is comparable to that obtained with software based DNN.</p></div>","PeriodicalId":21909,"journal":{"name":"Solid-state Electronics","volume":null,"pages":null},"PeriodicalIF":1.4000,"publicationDate":"2024-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Solid-state Electronics","FirstCategoryId":"101","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0038110124000091","RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0
Abstract
Deep learning has shown impressive capabilities in tasks like speech recognition and image classification. However, modern deep neural networks often demand a significant number of weights and extensive computational resources, creating efficiency challenges for applications on edge devices. To address these issues, researchers have introduced deep spiking neural networks (DSNNs) that leverage specialized hardware for synapses and neurons. DSNNs offer a potential solution by improving efficiency in edge-device applications. In this paper, the hardware based DSNN with integrate and fire neuron using steep switching device was investigated. We propose integrate and fire neuron using steep switching device to implement rate coding as input encoding method. Because the steep switching device has double-gate, the threshold voltage of the neuron circuits can be adaptively controlled, which changes the rates of input pulse. Hence, the adjustment of the threshold of neuron can be employed to mitigate the accuracy deterioration resulting from the transformation from deep neural networks (DNNs) to DSNNs. In addition, the off-current of proposed integrate and fire neuron circuit decreases significantly as the steep switching device has steep subthreshold swing. A system simulation of a hardware based DSNN shows that the adjustable threshold of the neuron circuit can achieve a high inference accuracy of 98.36 % which is comparable to that obtained with software based DNN.
期刊介绍:
It is the aim of this journal to bring together in one publication outstanding papers reporting new and original work in the following areas: (1) applications of solid-state physics and technology to electronics and optoelectronics, including theory and device design; (2) optical, electrical, morphological characterization techniques and parameter extraction of devices; (3) fabrication of semiconductor devices, and also device-related materials growth, measurement and evaluation; (4) the physics and modeling of submicron and nanoscale microelectronic and optoelectronic devices, including processing, measurement, and performance evaluation; (5) applications of numerical methods to the modeling and simulation of solid-state devices and processes; and (6) nanoscale electronic and optoelectronic devices, photovoltaics, sensors, and MEMS based on semiconductor and alternative electronic materials; (7) synthesis and electrooptical properties of materials for novel devices.