Hardware-Based Spiking Neural Networks Using Capacitor-Less Positive Feedback Neuron Devices
| Autor: | Jong-Ho Lee, Jong-Ho Bae, Dongseok Kwon, Byung-Gook Park, Suhwan Lim, Sung Yun Woo |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Spiking neural network
Artificial neural network business.industry Computer science Activation function Biological neuron model Subthreshold slope Electronic Optical and Magnetic Materials law.invention Capacitor law Electrical and Electronic Engineering business Computer hardware MNIST database Positive feedback |
| Zdroj: | IEEE Transactions on Electron Devices. 68:4766-4772 |
| ISSN: | 1557-9646 0018-9383 |
| DOI: | 10.1109/ted.2021.3098503 |
| Popis: | In this article, hardware-based spiking neural networks (SNNs) using capacitor-less positive feedback (PF) neuron devices are designed. It was reported that the PF device can simultaneously process the excitatory and inhibitory signals. The PF device shows very steep subthreshold slope (SS < 1 mV/dec) due to the PF opertaion, leading to low-power and reliable neuron device. The PF devices also show the behavior of leaky integrate and fire (LIF) neuron, which is the most popular neuron model in SNNs. For hardware configuration, the neuron characteristics of PF device are investigated with the transient behavior of the anode current. Based on the PF neuron devices, the SNN shows the accuracy of 98.19% for the Modified National Institute of Standards and Technology (MNIST) database classification in four-hidden layer, fully-connected neural network, which is near the accuracy (98.46%) of the artificial neural networks using rectified linear unit (ReLU) activation function. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|