4.6 Article

Intrinsic variation effect in memristive neural network with weight quantization

Related references

Note: Only part of the references are listed.
Article Engineering, Electrical & Electronic

Incremental Drain-Voltage-Ramping Training Method for Ferroelectric Field-Effect Transistor Synaptic Devices

Manh-Cuong Nguyen et al.

Summary: In this study, a HZO ferroelectric field-effect transistor (FeFET) was fabricated on a silicon-on-insulator substrate for MHz synaptic device applications. The FeFETs demonstrated stable multistate weights, excellent retention, and symmetric potentiation/depression (P/D). A novel incremental drain-voltage-ramping method was proposed to improve the linearity and symmetry of the P/D, and its compatibility was thoroughly verified. The results showed that the FeFETs achieved a linear and symmetric P/D with stable repeatability under a wide range of operating conditions, and achieved a learning accuracy of 95% in MNIST pattern recognition simulations.

IEEE ELECTRON DEVICE LETTERS (2022)

Article Mathematics, Interdisciplinary Applications

Effect of weight overlap region on neuromorphic system with memristive synaptic devices

Geun Ho Lee et al.

Summary: This study analyzes the effect of the conductance overlap region of memristors on the recognition accuracy for on-chip learning simulation.

CHAOS SOLITONS & FRACTALS (2022)

Article Mathematics, Interdisciplinary Applications

Conduction mechanism effect on physical unclonable function using Al2O3/TiOX memristors

Jinwoo Park et al.

Summary: This study evaluates the performance of a physical unclonable function (PUF) using Al2O3/TiOx based memristors. The wider distribution of high-resistance state (HRS) than low-resistance state (LRS) in PUF is confirmed through a conduction mechanism analysis. Additionally, cycle-to-cycle variation and random telegraph noise (RTN) characteristics impacting the reliability of the PUF are analyzed.

CHAOS SOLITONS & FRACTALS (2021)

Article Mathematics, Interdisciplinary Applications

Multilevel switching memristor by compliance current adjustment for off-chip training of neuromorphic system

Tae-Hyeon Kim et al.

Summary: This study explored the feasibility of achieving a larger number of conductance states using the I-cc control method, with experimental results demonstrating 64-level conductance states. Through off-chip learning of the CNN structure, it was verified that the 64-level states showed recognition performance close to that of software-based neural networks. By reducing information loss in the transfer process, the fabricated synaptic device array with the I-cc control programming method is expected to contribute to the development of hardware neural networks.

CHAOS SOLITONS & FRACTALS (2021)

Article Multidisciplinary Sciences

Spontaneous sparse learning for PCM-based memristor neural networks

Dong-Hyeok Lim et al.

Summary: The research team constructed a 39nm 1Gb phase-change memory memristor array and developed a spontaneous sparse learning scheme to improve PCM-based memristor network training. Experimental results show that this method helps enhance the performance and sparsity controllability of the network without requiring additional computation.

NATURE COMMUNICATIONS (2021)

Article Materials Science, Multidisciplinary

Controlled multilevel switching and artificial synapse characteristics in transparent HfAlO-alloy based memristor with embedded TaN nanoparticles

Chandreswar Mahata et al.

Summary: Atomic layer deposition was used to prepare tantalum nitride nanoparticles (TaN-NPs) for ITO/HfAlO/TaN-NP/HfAlO/ITO RRAM device. The presence of TaN-NPs improved device reliability and achieved multilevel conductance, with stable data retention by controlling RESET voltage.

JOURNAL OF MATERIALS SCIENCE & TECHNOLOGY (2021)

Article Engineering, Electrical & Electronic

Technological Design of 3D NAND-Based Compute-in-Memory Architecture for GB-Scale Deep Neural Network

Wonbo Shim et al.

Summary: This work proposes a heterogeneous integration strategy of 3DNAND based compute-in-memory (CIM) architecture for large-scale deep neural networks, aiming at language translation models with GB-scale parameters. By utilizing wafer bonding scheme and CMOS under array (CUA), the design integrates CMOS circuits, 3D NAND cells, and high voltage (HV) transistors at different tiers effectively. The energy efficiency achieved is 1.15-19.01 tera operations per second per watt (TOPS/W) using standard 3D NAND cells.

IEEE ELECTRON DEVICE LETTERS (2021)

Article Computer Science, Artificial Intelligence

A memristor-based circuit design of pavlov associative memory with secondary conditional reflex and its application

Sichun Du et al.

Summary: This paper proposes a Pavlov associative memory circuit with secondary conditional reflex utilizing memristor, which can respond to conditional stimulus after initial learning and have two kinds of forgetting process. It indirectly establishes conditioned reflexes through conditional stimuli and suggests an extended classification circuit based on secondary conditioned reflex.

NEUROCOMPUTING (2021)

Article Computer Science, Artificial Intelligence

Hardware -based spiking neural network architecture using simplified backpropagation algorithm and homeostasis functionality

Jangsaeng Kim et al.

Summary: The study proposes a supervised on-chip training method for hardware-based SNNs, achieving high performance through the use of pulse schemes and bias synapse. Experimental results show that the system achieves a recognition rate similar to that of software-based networks in MNIST dataset classification.

NEUROCOMPUTING (2021)

Article Nanoscience & Nanotechnology

3-bit multilevel operation with accurate programming scheme in TiO x /Al2O3 memristor crossbar array for quantized neuromorphic system

Tae-Hyeon Kim et al.

Summary: As interest in artificial intelligence and hardware technologies grows, the complexity of algorithms and network structures has led to significant power consumption issues. Neuromorphic computing, a hardware AI technology, aims to address this problem by focusing on multilevel operations and weight transfer methods for synaptic devices. Research has shown promise in achieving high classification accuracy and precise weight transfer in neuromorphic systems.

NANOTECHNOLOGY (2021)

Article Computer Science, Artificial Intelligence

A reconfigurable bidirectional associative memory network with memristor bridge

Yingying Li et al.

Summary: This study introduces a novel memristive bidirectional associative memory (MBAM) network circuit based on the threshold memristor bridge circuit (T-MBC), which shows excellent flexibility and reconfigurability. Experimental results demonstrate that the MBAM neural network achieves high recognition accuracy and strong noise immunity.

NEUROCOMPUTING (2021)

Article Engineering, Electrical & Electronic

Reconfigurable field-programmable gate array-based on-chip learning neuromorphic digital implementation for nonlinear function approximation

Morteza Gholami et al.

Summary: Hardware implementations of spiking neural networks, known as neuromorphic architectures, provide insight into brain performance and may inspire the next generation of computers and electronic systems. This paper discusses using Izhikevich neurons and gradient descent learning algorithm for approximating the sigmoid and other nonlinear functions, demonstrating the flexibility of spiking networks. Additionally, a time- and cost-efficient digital neuromorphic implementation method for approximating the sigmoid function is reported, with hardware synthesis results showing a maximum frequency of 83.209 MHz and throughput of 9.86 Mb/s.

INTERNATIONAL JOURNAL OF CIRCUIT THEORY AND APPLICATIONS (2021)

Article Multidisciplinary Sciences

4K-memristor analog-grade passive crossbar circuit

H. Kim et al.

Summary: The superior density of passive analog-grade memristive crossbar circuits allows for storing large neural network models on specialized neuromorphic chips to avoid costly off-chip communication. A 64x64 passive crossbar circuit with high functional nonvolatile metal-oxide memristors was successfully fabricated with low variation in memristor switching voltages. Experimental verification of analog properties demonstrated accurate modeling for MNIST image classification and large-scale multilayer perceptron classifier based on advanced conductance tuning algorithm.

NATURE COMMUNICATIONS (2021)

Article Chemistry, Physical

Ferroelectric polymer-based artificial synapse for neuromorphic computing

Sungjun Kim et al.

Summary: This study optimized the performance of ferroelectric materials by investigating the effects of different formation temperatures and contact metals on ferroelectric transistors, making breakthroughs in simulating synaptic characteristics. Metals forming relatively high barriers are beneficial for improving dynamic range and nonlinearity.

NANOSCALE HORIZONS (2021)

Article Computer Science, Artificial Intelligence

Efficient precise weight tuning protocol considering variation of the synaptic devices and target accuracy

Hyeongsu Kim et al.

NEUROCOMPUTING (2020)

Article Computer Science, Information Systems

3-D Synapse Array Architecture Based on Charge-Trap Flash Memory for Neuromorphic Application

Hyun-Seok Choi et al.

ELECTRONICS (2020)

Article Engineering, Electrical & Electronic

Fabrication and Characterization of TiOx Memristor for Synaptic Device Application

Tae-Hyeon Kim et al.

IEEE TRANSACTIONS ON NANOTECHNOLOGY (2020)

Article Chemistry, Physical

RRAM-based synapse devices for neuromorphic systems

K. Moon et al.

FARADAY DISCUSSIONS (2019)

Article Computer Science, Hardware & Architecture

Brain Intelligence: Go beyond Artificial Intelligence

Huimin Lu et al.

MOBILE NETWORKS & APPLICATIONS (2018)

Article Engineering, Biomedical

An On-Chip Learning Neuromorphic Autoencoder With Current-Mode Transposable Memory Read and Virtual Lookup Table

Hwasuk Cho et al.

IEEE TRANSACTIONS ON BIOMEDICAL CIRCUITS AND SYSTEMS (2018)

Review Physics, Applied

Recent progress in analog memory-based accelerators for deep learning

Hsinyu Tsai et al.

JOURNAL OF PHYSICS D-APPLIED PHYSICS (2018)

Article Engineering, Electrical & Electronic

Stuck-at Fault Tolerance in RRAM Computing Systems

Lixue Xia et al.

IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS (2018)

Article Chemistry, Multidisciplinary

Logic Computing with Stateful Neural Networks of Resistive Switches

Zhong Sun et al.

ADVANCED MATERIALS (2018)

Article Computer Science, Artificial Intelligence

Sign backpropagation: An on-chip learning algorithm for analog RRAM neuromorphic computing systems

Qingtian Zhang et al.

NEURAL NETWORKS (2018)

Article Chemistry, Physical

Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing

Zhongrui Wang et al.

NATURE MATERIALS (2017)

Article Acoustics

Multichannel Signal Processing With Deep Neural Networks for Automatic Speech Recognition

Tara N. Sainath et al.

IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING (2017)

Article Nanoscience & Nanotechnology

Analog Synaptic Behavior of a Silicon Nitride Memristor

Sungjun Kim et al.

ACS APPLIED MATERIALS & INTERFACES (2017)

Review Physics, Multidisciplinary

Neuromorphic computing using non-volatile memory

Geoffrey W. Burr et al.

ADVANCES IN PHYSICS-X (2017)

Article Computer Science, Hardware & Architecture

Technological Exploration of RRAM Crossbar Array for Matrix-Vector Multiplication

Lixue Xia et al.

JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY (2016)

Article Computer Science, Software Engineering

Latte: A Language, Compiler, and Runtime for Elegant and Efficient Deep Neural Networks

Leonard Truong et al.

ACM SIGPLAN NOTICES (2016)

Article Engineering, Electrical & Electronic

Deep Neural Network Approaches to Speaker and Language Recognition

Fred Richardson et al.

IEEE SIGNAL PROCESSING LETTERS (2015)

Article Computer Science, Hardware & Architecture

True North: Design and Tool Flow of a 65 mW 1 Million Neuron Programmable Neurosynaptic Chip

Filipp Akopyan et al.

IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS (2015)

Article Computer Science, Hardware & Architecture

RRAM Defect Modeling and Failure Analysis Based on March Test and a Novel Squeeze-Search Scheme

Ching-Yi Chen et al.

IEEE TRANSACTIONS ON COMPUTERS (2015)

Article Automation & Control Systems

Neuromorphic Hardware System for Visual Pattern Recognition With Memristor Array and CMOS Neuron

Myonglae Chu et al.

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS (2015)

Article Engineering, Electrical & Electronic

Causes and consequences of the stochastic aspect of filamentary RRAM

R. Degraeve et al.

MICROELECTRONIC ENGINEERING (2015)

Article Nanoscience & Nanotechnology

Integration of nanoscale memristor synapses in neuromorphic computing architectures

Giacomo Indiveri et al.

NANOTECHNOLOGY (2013)

Article Multidisciplinary Sciences

Pattern classification by memristive crossbar circuits using ex situ and in situ training

Fabien Alibart et al.

NATURE COMMUNICATIONS (2013)

Article Engineering, Electrical & Electronic

Visual Pattern Extraction Using Energy-Efficient 2-PCM Synapse Neuromorphic Architecture

Olivier Bichler et al.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2012)