☆ 4.6 Article

Super-steep synapses based on positive feedback devices for reliable binary neural networks

APPLIED PHYSICS LETTERS (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Engineering, Electrical & Electronic

Demonstration of Pulse Width Modulation Function Using Single Positive Feedback Device for Neuron

Sung Yun Woo et al.

Summary: A positive-feedback (PF) device and its operation scheme for implementing pulse width modulation (PWM) function were demonstrated. By adjusting the charge stored in the n(-) floating body (Q(n)), the potential of the floating body changes linearly with time. The voltage-to-pulse width conversion for PWM function was achieved by utilizing the linear time-varying property of Q(n) and the gate bias dependency of Q(th). This approach allows for the reduction of the area required for a PWM neuron.

IEEE ELECTRON DEVICE LETTERS (2023)

添加到收藏夹

Article Physics, Applied

Dual-configuration in-memory computing bitcells using SiOx RRAM for binary neural networks

Sandeep Kaur Kingra et al.

Summary: Conventional DNN implementations face limitations in the context of edge-AI due to high precision computation requirements, large memory demand, and memory wall. To address this, low-precision DNN implementations based on in-memory computing (IMC) using non-volatile memory (NVM) have been explored. This work presents the experimental demonstration of a dual-configuration XNOR IMC bitcell and its functionality in binarized neural networks (BNNs).

APPLIED PHYSICS LETTERS (2022)

添加到收藏夹

Article Engineering, Electrical & Electronic

SiO2 Fin-Based Flash Synaptic Cells in AND Array Architecture for Binary Neural Networks

Soochang Lee et al.

Summary: An oxide fin-based AND flash memory synaptic device is proposed and fabricated for a hardware-based binary neural network. The device features a fin-like curved channel structure, providing local electric field enhancement for improved programming efficiency. The fin-based AND flash cell exhibits high on/off current ratio, low off current, and low programming voltage, resulting in a sufficient dynamic range of synaptic weights for BNNs. Additionally, a hardware-based BNN using a novel two-cell synaptic device arrangement is proposed, enabling parallel XNOR operation and bit-counting. Experimental results show only slight degradation in classification accuracy compared to the baseline accuracy, making it suitable for off-chip event-driven computation.

IEEE ELECTRON DEVICE LETTERS (2022)

添加到收藏夹

Article Engineering, Electrical & Electronic

Variation-Tolerant Capacitive Array for Binarized Neural Network

Hyeongsu Kim et al.

Summary: This letter proposes a variation-tolerant capacitive binarized neural network using 2-terminal MANOS memory devices. The capacitance-voltage characteristic of the synaptic device is used to represent the weight of the binarized neural network. The proposed synaptic device shows less capacitance variation compared to conventional synaptic devices, resulting in a more robust network.

IEEE ELECTRON DEVICE LETTERS (2022)

添加到收藏夹

Article Engineering, Electrical & Electronic

An Accurate, Error-Tolerant, and Energy-Efficient Neural Network Inference Engine Based on SONOS Analog Memory

T. Patrick Xiao et al.

Summary: We demonstrate SONOS analog memory arrays optimized for neural network inference, achieving high accuracy and energy efficiency. With subthreshold operation, low conductances can be implemented with low error, matching the weight distribution of neural networks. The system achieves accuracy within 2.16% of floating-point accuracy and a >10x gain in energy efficiency over state-of-the-art accelerators.

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS (2022)

添加到收藏夹

Review Engineering, Electrical & Electronic

Review of Analog Neuron Devices for Hardware-based Spiking Neural Networks

Dongseok Kwon et al.

Summary: In order to process data operations more efficiently in deep neural networks, researchers have been studying spiking neural networks. CMOS neuron circuits that mimic the biological behavior of neurons have been the main focus in the reported literature. However, conventional neuronal circuits need improvement in terms of area and energy consumption, leading to the emergence of neuron devices with memory functions. In this review article, neuron devices that can increase integration density and reduce power consumption in conventional neuronal circuits are discussed, and their potential importance in future neuromorphic systems is emphasized.

JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE (2022)

添加到收藏夹

Article Chemistry, Analytical

Efficient fusion of spiking neural networks and FET-type gas sensors for a fast and reliable artificial olfactory system

Dongseok Kwon et al.

Summary: A new artificial olfactory system based on a spiking neural network (SNN) and field-effect transistor (FET)-type gas sensors was proposed for quickly and reliably detecting toxic gases. The system showed a low error rate in predicting the concentrations of NO2 and H2S, with real-time gas concentration prediction achieved within approximately 5 seconds. The hardware-based SNN outperformed conventional machine learning algorithms, showcasing its superiority in detecting toxic gases.

SENSORS AND ACTUATORS B-CHEMICAL (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

3-D AND-Type Flash Memory Architecture With High-κ Gate Dielectric for High-Density Synaptic Devices

Young-Tak Seo et al.

Summary: This paper introduces a novel 3-D synaptic device structure with features of high-speed operation, high density, and stacking layers, and successfully verifies its potential application in neuromorphic networks. Additionally, hardware-based CNN design and accuracy evaluation of CIFAR-10 image classification are demonstrated.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2021)

添加到收藏夹

Article Computer Science, Information Systems

Maximizing Parallel Activation of Word-Lines in MRAM-Based Binary Neural Network Accelerators

Daehyun Ahn et al.

Summary: A proposed algorithm/hardware co-design methodology improves the tolerance of Binary Neural Networks (BNNs) to Magnetic RAM (MRAM) variation by selecting a memristor crossbar array (MCA) structure with good sensitivity to process variations, applying input distribution regularization scheme, and adopting retraining scheme based on knowledge distillation. This approach significantly increases the number of parallel word-line activations while enhancing throughput and energy efficiency.

IEEE ACCESS (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

An Improved RRAM-Based Binarized Neural Network With High Variation-Tolerated Forward/Backward Propagation Module

Yizhou Zhang et al.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2020)

添加到收藏夹

Article Computer Science, Hardware & Architecture

Accelerating Deep Neural Network In-Situ Training With Non-Volatile and Volatile Memory Based Hybrid Precision Synapses

Yandong Luo et al.

IEEE TRANSACTIONS ON COMPUTERS (2020)

添加到收藏夹

Article Neurosciences

On-Chip Training Spiking Neural Networks Using Approximated Backpropagation With Analog Synaptic Devices

Dongseok Kwon et al.

FRONTIERS IN NEUROSCIENCE (2020)

添加到收藏夹

Article Physics, Applied

Machine learning method to predict threshold voltage distribution by read disturbance in 3D NAND Flash Memories

Jihye Park et al.

JAPANESE JOURNAL OF APPLIED PHYSICS (2020)

添加到收藏夹

Article Engineering, Electrical & Electronic

Reliability of NAND Flash Memory as a Weight Storage Device of Artificial Neural Network

Md Mehedi Hasan et al.

IEEE TRANSACTIONS ON DEVICE AND MATERIALS RELIABILITY (2020)

添加到收藏夹

Article Engineering, Electrical & Electronic

High-Throughput In-Memory Computing for Binary Deep Neural Networks With Monolithically Integrated RRAM and 90-nm CMOS

Shihui Yin et al.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2020)

添加到收藏夹

Article Engineering, Electrical & Electronic

Resistive Crossbars as Approximate Hardware Building Blocks for Machine Learning: Opportunities and Challenges

Indranil Chakraborty et al.

PROCEEDINGS OF THE IEEE (2020)

添加到收藏夹

Article Computer Science, Information Systems

Low-Power Binary Neuron Circuit With Adjustable Threshold for Binary Neural Networks Using NAND Flash Memory

Sung-Tae Lee et al.

IEEE ACCESS (2020)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Multimodal Machine Learning: A Survey and Taxonomy

Tadas Baltrusaitis et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2019)

添加到收藏夹

Article Computer Science, Hardware & Architecture

PXNOR-BNN: In/With Spin-Orbit Torque MRAM Preset-XNOR Operation-Based Binary Neural Networks

Liang Chang et al.

IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS (2019)

添加到收藏夹

Editorial Material Engineering, Electrical & Electronic

Change of Editor-in-Chief

Hisayo S. Momose et al.

IEEE ELECTRON DEVICE LETTERS (2019)

添加到收藏夹

Article Multidisciplinary Sciences

Equivalent-accuracy accelerated neural-network training using analogue memory

Stefano Ambrogio et al.

NATURE (2018)

添加到收藏夹

Article Engineering, Electrical & Electronic

Neuro-Inspired Computing With Emerging Nonvolatile Memory

Shimeng Yu

PROCEEDINGS OF THE IEEE (2018)

添加到收藏夹

Article Physics, Applied

Integrate-and-fire neuron circuit using positive feedback field effect transistor for low power operation

Min-Woo Kwon et al.

JOURNAL OF APPLIED PHYSICS (2018)

添加到收藏夹

Article Engineering, Electrical & Electronic

Charge Loss Mechanisms of Nitride-Based Charge Trap Flash Memory Devices

Meng Chuan Lee et al.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2013)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.