Quantized STDP-based online-learning spiking neural network

Hu, Shaogang; Qiao, Guanchao; Chen, T. P.; Yu, Qin; Liu, Y.; Rong, Liang

doi:10.1007/s00521-021-05832-y

Cited by 25 publications

(10 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…η post /η pre represent the pre- and postsynaptic learning rates. To mimic quantized weight updates typically observed in such device technologies ( 49 ), the weights in the network-level simulations were quantized usingwq=roundfalse[false(2N−1false)×wfalse]2N−1…”

Section: Methodsmentioning

confidence: 99%

Selective area doping for Mott neuromorphic electronics

Deng

Park

et al. 2023

Sci. Adv.

View full text Add to dashboard Cite

The cointegration of artificial neuronal and synaptic devices with homotypic materials and structures can greatly simplify the fabrication of neuromorphic hardware. We demonstrate experimental realization of vanadium dioxide (VO 2 ) artificial neurons and synapses on the same substrate through selective area carrier doping. By locally configuring pairs of catalytic and inert electrodes that enable nanoscale control over carrier density, volatility or nonvolatility can be appropriately assigned to each two-terminal Mott memory device per lithographic design, and both neuron- and synapse-like devices are successfully integrated on a single chip. Feedforward excitation and inhibition neural motifs are demonstrated at hardware level, followed by simulation of network-level handwritten digit and fashion product recognition tasks with experimental characteristics. Spatially selective electron doping opens up previously unidentified avenues for integration of emerging correlated semiconductors in electronic device technologies.

show abstract

Section: Methodsmentioning

confidence: 99%

Selective area doping for Mott neuromorphic electronics

Deng

Park

et al. 2023

Sci. Adv.

View full text Add to dashboard Cite

show abstract

“…The most common model to train SNN is called synaptic timedependent plasticity (STDP) unsupervised approach [26]. The STDP is utilized along with the side restrained fit spiking threshold to learn exemplification for input spike paradigms which are appropriate for classification [27]. The spikes are encoded by converting the input wave signal into a sequence of spikes "spiketrains" in a process called "encoding".…”

Section: Issn: 2088-8708 mentioning

confidence: 99%

Intrusion detection method for internet of things based on the spiking neural network and decision tree method

Zarzoor¹,

Al-Jamali

Qader

2023

IJECE

View full text Add to dashboard Cite

The prevalence of using the applications for the internet of things (IoT) in many human life fields such as economy, social life, and healthcare made IoT devices targets for many cyber-attacks. Besides, the resource limitation of IoT devices such as tiny battery power, small storage capacity, and low calculation speed made its security a big challenge for the researchers. Therefore, in this study, a new technique is proposed called intrusion detection system based on spike neural network and decision tree (IDS-SNNDT). In this method, the DT is used to select the optimal samples that will be hired as input to the SNN, while SNN utilized the non-leaky integrate neurons fire (NLIF) model in order to reduce latency and minimize devices’ power usage. Also, a rand order code (ROC) technique is used with SNN to detect cyber-attacks. The proposed method is evaluated by comparing its performance with two other methods: IDS-DNN and IDS-SNNTLF by using three performance metrics: detection accuracy, latency, and energy usage. The simulation results have shown that IDS-SNNDT attained low power usage and less latency in comparison with IDS-DNN and IDS-SNNTLF methods. Also, IDS-SNNDT has achieved high detection accuracy for cyber-attacks in contrast with IDS-SNNTLF.

show abstract

“…In reality, from a system design perspective, we need to have interleaved synaptic device state update phases that do not interfere with the neuron oscillation behavior (for instance, through decoupled write-read phases of three-terminal synaptic devices; Sengupta et al, 2016 ). The convergence was also not affected with reduced programming resolution of the synaptic connections (4-bits), thereby indicating resiliency to quantization (Hu et al, 2021 ).…”

Section: Binding Problemmentioning

confidence: 99%

Emulation of Astrocyte Induced Neural Phase Synchrony in Spin-Orbit Torque Oscillator Neurons

2021

View full text Add to dashboard Cite

Astrocytes play a central role in inducing concerted phase synchronized neural-wave patterns inside the brain. In this article, we demonstrate that injected radio-frequency signal in underlying heavy metal layer of spin-orbit torque oscillator neurons mimic the neuron phase synchronization effect realized by glial cells. Potential application of such phase coupling effects is illustrated in the context of a temporal “binding problem.” We also present the design of a coupled neuron-synapse-astrocyte network enabled by compact neuromimetic devices by combining the concepts of local spike-timing dependent plasticity and astrocyte induced neural phase synchrony.

show abstract

Quantized STDP-based online-learning spiking neural network

Cited by 25 publications

References 40 publications

Selective area doping for Mott neuromorphic electronics

Selective area doping for Mott neuromorphic electronics

Intrusion detection method for internet of things based on the spiking neural network and decision tree method

Emulation of Astrocyte Induced Neural Phase Synchrony in Spin-Orbit Torque Oscillator Neurons

Contact Info

Product

Resources

About