Memristor-based synaptic plasticity and unsupervised learning of spiking neural networks

Hajiabadi, Zohreh; Shalchian, Majid

doi:10.1007/s10825-021-01719-2

Cited by 6 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, an ANN that can be used for neuromorphic computing can be constructed through a crossbar array based on nonvolatile memristive devices. At present, hardware ANNs based on various neural networks have been successfully applied in pattern recognition. , For instance, SNNs based on the STDP learning rule can realize pattern recognition. Specifically, the synaptic weights in SNN are modulated according to the difference in the spike timing of the input-layer and output-layer neurons.…”

Section: Ann and Rc Based On Memristive Devicesmentioning

confidence: 99%

2D-Material-Based Volatile and Nonvolatile Memristive Devices for Neuromorphic Computing

Xia

Huang

Hang

et al. 2023

ACS Materials Lett.

View full text Add to dashboard Cite

Neuromorphic computing can process large amounts of information in parallel and provides a powerful tool to solve the von Neumann bottleneck. Constructing an artificial neural network (ANN) is a common means to realize neuromorphic computing, which has exhibited potential applications in pattern recognition, complex sensing, and other areas. Reservoir computing (RC), which is another approach to realize neuromorphic computing, has shown some progress and attracted researchers' attention. Neuromorphic computing can be generally implemented by fabricating memristive array systems. 2D-material-based memristive systems and their applications in ANN and RC have been investigated substantially in recent years due to the unique properties of these systems, such as atomic-level thickness and high carrier mobility. In this Review, we first discuss the volatility and nonvolatility properties of memristive devices and their applications in ANN and RC. Second, 2D materials that can be used to fabricate these devices are introduced, and their classification, physical properties, and preparation methods are presented. Third, we discuss the working mechanisms of 2D-material-based synaptic devices, the mimicked synaptic functions, and the applications of these devices in neuromorphic computing through ANN and RC. Lastly, the performance, progress, and future development directions of 2D-material-based synaptic devices are analyzed. This work systematically investigates the status of 2D-material-based synaptic devices and promotes their utilization in neuromorphic computing.

show abstract

Section: Ann and Rc Based On Memristive Devicesmentioning

confidence: 99%

2D-Material-Based Volatile and Nonvolatile Memristive Devices for Neuromorphic Computing

Xia

Huang

Hang

et al. 2023

ACS Materials Lett.

View full text Add to dashboard Cite

show abstract

“…The investigations on braininspired SNN have been increasing sharply with high parallelism and higher efficiency [3,4], which is inspired by the efficient human brain. The operational mechanism of the human brain has served as a significant inspiration for numerous researchers, prompting their focus on spike-timing-dependent plasticity (STDP) learning rules that exhibit greater biological plausibility [5]. The investigation of artificial synapses and neurons is crucial for the advancement of hardware implementation in neuromorphic systems relying on spiking neural networks (SNNs) [6,7].…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Learning of Digit Recognition Through Spike-Timing-Dependent Plasticity Based on Memristors

Wang,

Yan,

Liu

et al. 2023

Atlantis Highlights in Computer Sciences

View full text Add to dashboard Cite

Neuromorphic computing based on spiking neural networks (SNNs) is a promising alternative in the field of intelligent computing, especially when traditional Von Neumann architectures is facing several choke point. Memristors, as the fourth-generation fundamental circuit element, play a crucial role in neuromorphic computing systems and are commonly employed as neural and synaptic devices. Due to their spike-based operation, memristive spiking neural networks (MSNNs) are considered to be superior and biologically plausible compared to alternative systems in terms of effectiveness. Here, the spike-timing-dependent plasticity (STDP) learning characteristic is reaped from our manufactured equipment. Utilizing memristor-based leaky integrate-and-fire (LIF) neurons and synapses, unsupervised learning of spiking neural networks with 784 × 324 × 324 architectures are constructed.

show abstract

“…Zayer et al [18] present the implementation of low power and ultrafast spiketiming-dependent plasticity (STDP) of the spiking neural network (SNN) in a crossbar structure based on the ferroelectric tunnel memristor (FTM). Zohreh et al [19] used LIF neurons and memristive synapses to construct a fully connected SNN with 2 × 2 and 4 × 2 structures for pattern classification. In these works, memristors are used as synapses of neural networks, demonstrating the value of memristors in neural networks.…”

Section: Introductionmentioning

confidence: 99%

Memristor-based multi-synaptic spiking neuron circuit for spiking neural network

Jiang

Liu

et al. 2022

Chinese Phys. B

View full text Add to dashboard Cite

Spiking neural networks (SNNs) are widely used in many fields because they work closer to biological neurons. However, due to its computational complexity, many SNNs implementations are limited to computer programs. First, this paper proposes a multi-synaptic circuit (MSC) based on memristor, which realizes the multi-synapse connection between neurons and the multi-delay transmission of pulse signals. The synapse circuit participates in the calculation of the network while transmitting the pulse signal, and completes the complex calculations on the software with hardware. Secondly, a new spiking neuron circuit based on the leaky integrate-and-fire (LIF) model is designed in this paper. The amplitude and width of the pulse emitted by the spiking neuron circuit can be adjusted as required. The combination of spiking neuron circuit and MSC forms the multi-synaptic spiking neuron (MSSN). The MSSN was simulated in PSPICE and the expected result was obtained, which verified the feasibility of the circuit. Finally, a small SNN was designed based on the mathematical model of MSSN. After the SNN is trained and optimized, it obtains a good accuracy in the classification of the IRIS-dataset, which verifies the practicability of the design in the network.

show abstract

Memristor-based synaptic plasticity and unsupervised learning of spiking neural networks

Cited by 6 publications

References 23 publications

2D-Material-Based Volatile and Nonvolatile Memristive Devices for Neuromorphic Computing

2D-Material-Based Volatile and Nonvolatile Memristive Devices for Neuromorphic Computing

Unsupervised Learning of Digit Recognition Through Spike-Timing-Dependent Plasticity Based on Memristors

Memristor-based multi-synaptic spiking neuron circuit for spiking neural network

Contact Info

Product

Resources

About