A Review of Computing with Spiking Neural Networks

Artificial neural networks(ANNs)have led to landmark changes in many fields,but they still differ significantly fromthemechanisms of real biological neural networks and face problems such as high computing costs,excessive computing power,and so on.Spiking neural networks(SNNs)provide a new approach combined with brain-like science to improve the computational energy efficiency,computational architecture,and biological credibility of current deep learning applications.In the early stage of development,its poor performance hindered the application of SNNs in real-world scenarios.In recent years,SNNs have made great progress in computational performance and practicability compared with the earlier research results,and are continuously producing significant results.Although there are already many pieces of literature on SNNs,there is still a lack of comprehensive review on SNNs from the perspective of improving performance and practicality as well as incorporating the latest research results.Starting from this issue,this paper elaborates on SNNs along the complete usage process of SNNs including network construction,data processing,model training,development,and deployment,aiming to provide more comprehensive and practical guidance to promote the development of SNNs.Therefore,the connotation and development status of SNNcomputing is reviewed systematically and comprehensively from four aspects:composition structure,data set,learning algorithm,software/hardware development platform.Then the development characteristics of SNNs in intelligent computing are summarized,the current challenges of SNNs are discussed and the future development directions are also prospected.Our research shows that in the fields of machine learning and intelligent computing,SNNs have comparable network scale and performance to ANNs and the ability to challenge large datasets and a variety of tasks.The advantages of SNNs over ANNs in terms of energy efficiency and spatial-temporal data processing have been more fully exploited.And the development of programming and deployment tools has lowered the threshold for the use of SNNs.SNNs show a broad development prospect for brain-like computing.

Parallel architecture and optimization for discrete-event simulation of spike neural networks

Spike neural networks are inspired by animal brains,and outperform traditional neural networks on complicated tasks.However,spike neural networks are usually used on a large scale,and they cannot be computed on commercial,off-the-shelf computers.A parallel architecture is proposed and developed for discrete-event simulations of spike neural networks.Furthermore,mechanisms for both parallelism degree estimation and dynamic load balance are emphasized with theoretical and computational analysis.Simulation results show the effectiveness of the proposed parallelized spike neural network system and its corresponding support components.

Photonic integrated neuro-synaptic core for convolutional spiking neural network

Neuromorphic photonic computing has emerged as a competitive computing paradigm to overcome the bottlenecks of the von-Neumann architecture.Linear weighting and nonlinear spike activation are two fundamental functions of a photonic spiking neural network(PSNN).However,they are separately implemented with different photonic materials and devices,hindering the large-scale integration of PSNN.Here,we propose,fabricate and experimentally demonstrate a photonic neuro-synaptic chip enabling the simultaneous implementation of linear weighting and nonlinear spike activation based on a distributed feedback(DFB)laser with a saturable absorber(DFB-SA).A prototypical system is experimentally constructed to demonstrate the parallel weighted function and nonlinear spike activation.Furthermore,a fourchannel DFB-SA laser array is fabricated for realizing matrix convolution of a spiking convolutional neural network,achieving a recognition accuracy of 87%for the MNIST dataset.The fabricated neuro-synaptic chip offers a fundamental building block to construct the large-scale integrated PSNN chip.

Pattern recognition in multi-synaptic photonic spiking neural networks based on a DFB-SA chip

Spiking neural networks(SNNs)utilize brain-like spatiotemporal spike encoding for simulating brain functions.Photonic SNN offers an ultrahigh speed and power efficiency platform for implementing high-performance neuromorphic computing.Here,we proposed a multi-synaptic photonic SNN,combining the modified remote supervised learning with delayweight co-training to achieve pattern classification.The impact of multi-synaptic connections and the robustness of the network were investigated through numerical simulations.In addition,the collaborative computing of algorithm and hardware was demonstrated based on a fabricated integrated distributed feedback laser with a saturable absorber(DFB-SA),where 10 different noisy digital patterns were successfully classified.A functional photonic SNN that far exceeds the scale limit of hardware integration was achieved based on time-division multiplexing,demonstrating the capability of hardware-algorithm co-computation.

A progressive surrogate gradient learning for memristive spiking neural network

In recent years, spiking neural networks(SNNs) have received increasing attention of research in the field of artificial intelligence due to their high biological plausibility, low energy consumption, and abundant spatio-temporal information.However, the non-differential spike activity makes SNNs more difficult to train in supervised training. Most existing methods focusing on introducing an approximated derivative to replace it, while they are often based on static surrogate functions. In this paper, we propose a progressive surrogate gradient learning for backpropagation of SNNs, which is able to approximate the step function gradually and to reduce information loss. Furthermore, memristor cross arrays are used for speeding up calculation and reducing system energy consumption for their hardware advantage. The proposed algorithm is evaluated on both static and neuromorphic datasets using fully connected and convolutional network architecture, and the experimental results indicate that our approach has a high performance compared with previous research.

Deep Learning with Optimal Hierarchical Spiking Neural Network for Medical Image Classification

Medical image classification becomes a vital part of the design of computer aided diagnosis(CAD)models.The conventional CAD models are majorly dependent upon the shapes,colors,and/or textures that are problem oriented and exhibited complementary in medical images.The recently developed deep learning(DL)approaches pave an efficient method of constructing dedicated models for classification problems.But the maximum resolution of medical images and small datasets,DL models are facing the issues of increased computation cost.In this aspect,this paper presents a deep convolutional neural network with hierarchical spiking neural network(DCNN-HSNN)for medical image classification.The proposed DCNN-HSNN technique aims to detect and classify the existence of diseases using medical images.In addition,region growing segmentation technique is involved to determine the infected regions in the medical image.Moreover,NADAM optimizer with DCNN based Capsule Network(CapsNet)approach is used for feature extraction and derived a collection of feature vectors.Furthermore,the shark smell optimization algorithm(SSA)based HSNN approach is utilized for classification process.In order to validate the better performance of the DCNN-HSNN technique,a wide range of simulations take place against HIS2828 and ISIC2017 datasets.The experimental results highlighted the effectiveness of the DCNN-HSNN technique over the recent techniques interms of different measures.Please type your abstract here.

SpikeGoogle:Spiking Neural Networks with GoogLeNet-like inception module

Spiking Neural Network is known as the third-generation artificial neural network whose development has great potential.With the help of Spike Layer Error Reassignment in Time for error back-propagation,this work presents a new network called SpikeGoogle,which is implemented with GoogLeNet-like inception module.In this inception module,different convolution kernels and max-pooling layer are included to capture deep features across diverse scales.Experiment results on small NMNIST dataset verify the results of the authors’proposed SpikeGoogle,which outperforms the previous Spiking Convolutional Neural Network method by a large margin.

Efficient hybrid neural network for spike sorting

Artificial neural network has been used successfully to develope the automatic spike extraction. In order to address some of the problems before the wireless transmission of the implantable chip, the automatic spike sorting method with low complexity and high efficiency is proposed based on the hybrid neural network with the principal component analysis network （PCAN） and normal boundary response （NBR） self-organizing mapping （SOM） net- work classifier. An automatic PCAN technique is used to reduce the dimension and eliminate the correlation of the spike signal. The NBR-SOM network performs the spike sorting challenge and improves the classification performance. The experimental results show that based on the hybrid neural network, the spike sorting method achieves the accuracy above 97.91% with signals contain- ing five classes. The proposed NBR-SOM network classifier is to further improve the stability and effectiveness of the classification system.

Biological Neural Network Structure and Spike Activity Prediction Based on Multi-Neuron Spike Train Data

The micro-scale neural network structure for the brain is essential for the investigation on the brain and mind. Most of the previous studies typically acquired the neural network structure through brain slicing and reconstruction via nanoscale imaging. Nevertheless, this method still cannot scale well, and the observation on the neural activities based on the reconstructed neural network is not possible. Neuron activities are based on the neural network of the brain. In this paper, we propose that multi-neuron spike train data can be used as an alternative source to predict the neural network structure. And two concrete strategies for neural network structure prediction based on such kind of data are introduced, namely, the time-ordered strategy and the spike co-occurrence strategy. The proposed methods can even be applied to in vivo studies since it only requires neural spike activities. Based on the predicted neural network structure and the spreading activation theory, we propose a spike prediction method. For neural network structure reconstruction, the experimental results reveal a significantly improved accuracy compared to previous network reconstruction strategies, such as Cross-correlation, Pearson, and the Spearman method. Experiments on the spikes prediction results show that the proposed spreading activation based strategy is potentially effective for predicting neural spikes in the biological neural network. The predictions on the neural network structure and the neuron activities serve as foundations for large scale brain simulation and explorations of human intelligence.

Fast Learning in Spiking Neural Networks by Learning Rate Adaptation

For accelerating the supervised learning by the SpikeProp algorithm with the temporal coding paradigm in spiking neural networks (SNNs), three learning rate adaptation methods (heuristic rule, delta-delta rule, and delta-bar-delta rule), which are used to speed up training in artificial neural networks, are used to develop the training algorithms for feedforward SNN. The performance of these algorithms is investigated by four experiments: classical XOR (exclusive or) problem, Iris dataset, fault diagnosis in the Tennessee Eastman process, and Poisson trains of discrete spikes. The results demonstrate that all the three learning rate adaptation methods are able to speed up convergence of SNN compared with the original SpikeProp algorithm. Furthermore, if the adaptive learning rate is used in combination with the momentum term, the two modifications will balance each other in a beneficial way to accomplish rapid and steady convergence. In the three learning rate adaptation methods, delta-bar-delta rule performs the best. The delta-bar-delta method with momentum has the fastest convergence rate, the greatest stability of training process, and the maximum accuracy of network learning. The proposed algorithms in this paper are simple and efficient, and consequently valuable for practical applications of SNN.

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

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.

Integrated Evolving Spiking Neural Network and Feature Extraction Methods for Scoliosis Classification

Adolescent Idiopathic Scoliosis(AIS)is a deformity of the spine that affects teenagers.The current method for detecting AIS is based on radiographic images which may increase the risk of cancer growth due to radiation.Photogrammetry is another alternative used to identify AIS by distinguishing the curves of the spine from the surface of a human’s back.Currently,detecting the curve of the spine is manually performed,making it a time-consuming task.To overcome this issue,it is crucial to develop a better model that automatically detects the curve of the spine and classify the types of AIS.This research proposes a new integration of ESNN and Feature Extraction(FE)methods and explores the architecture of ESNN for the AIS classification model.This research identifies the optimal Feature Extraction(FE)methods to reduce computational complexity.The ability of ESNN to provide a fast result with a simplicity and performance capability makes this model suitable to be implemented in a clinical setting where a quick result is crucial.A comparison between the conventional classifier(Support Vector Machine(SVM),Multi-layer Perceptron(MLP)and Random Forest(RF))with the proposed AIS model also be performed on a dataset collected by an orthopedic expert from Hospital Universiti Kebangsaan Malaysia(HUKM).This dataset consists of various photogrammetry images of the human back with different types ofMalaysian AIS patients to solve the scoliosis problem.The process begins by pre-processing the images which includes resizing and converting the captured pictures to gray-scale images.This is then followed by feature extraction,normalization,and classification.The experimental results indicate that the integration of LBP and ESNN achieves higher accuracy compared to the performance of multiple baseline state-of-the-art Machine Learning for AIS classification.This demonstrates the capability of ESNN in classifying the types of AIS based on photogrammetry images.

基于脉冲神经网络的时空交互图像分类

脉冲神经网络作为人工智能发展的重要方向之一,在神经形态工程和类脑计算领域得到了广泛的关注.为解决脉冲神经网络泛化性差、内存和时间消耗较大等问题,本文提出了一种基于脉冲神经网络的时空交互图像分类方法.首先引入时间有效训练算法弥补梯度下降过程中的动能损失;其次融合空间随时间学习算法,提高网络对信息的高效处理能力;最后添加空间注意力机制,增强网络对空间维度上重要特征的捕捉能力.实验结果表明,改进后的方法在CIFAR10、DVS Gesture、CIFAR10-DVS这3个数据集上的训练内存占用分别减少了46.68%、48.52%、10.46%,训练速度分别提升了2.80倍、1.31倍、2.76倍,在保证精度的情况下,网络性能得到有效提升.

基于多维投影时空事件帧的动态视觉传感手势识别

基于视觉的手势识别是虚拟现实、游戏仿真等领域常用的人机交互手段。在实际应用中,手势动作快速变化将导致传统RGB相机或深度相机成像模糊,给手势识别带来巨大挑战。针对上述问题,利用动态视觉传感器捕捉高速手势运动信息,提出一种基于多维投影时空事件帧(spatiotemporal event frame,STEF)的动态视觉数据手势识别方法。将时空信息嵌入到数据投影面融合形成多维投影时空事件帧,克服现有动态视觉信息事件帧表达方法时域信息丢失的局限性,提升动态视觉传感数据的特征表达能力。在此基础上,采用先进的脉冲神经网络对时空事件帧进行分类实现手势识别。在公开数据集上的识别精度达到96.67%,性能优于同类方法,表明该方法可显著提升动态视觉传感数据手势识别准确率。

Memristor-based spiking neural networks:cooperative development of neural network architecture/algorithms and memristors

Inspired by the structure and principles of the human brain,spike neural networks(SNNs)appear as the latest generation of artificial neural networks,attracting significant and universal attention due to their remarkable low-energy transmission by pulse and powerful capability for large-scale parallel computation.Current research on artificial neural networks gradually change from software simulation into hardware implementation.However,such a process is fraught with challenges.In particular,memristors are highly anticipated hardware candidates owing to their fastprogramming speed,low power consumption,and compatibility with the complementary metal–oxide semiconductor(CMOS)technology.In this review,we start from the basic principles of SNNs,and then introduced memristor-based technologies for hardware implementation of SNNs,and further discuss the feasibility of integrating customized algorithm optimization to promote efficient and energy-saving SNN hardware systems.Finally,based on the existing memristor technology,we summarize the current problems and challenges in this field.

脉冲非对称卷积神经网络的图像与事件分类算法

为了提升模型性能的同时不引入额外的计算量与能量消耗,提出了一种脉冲非对称卷积算法。利用卷积核交叉部分的权重大的特点,采用多个尺寸的卷积核替换普通卷积的单个卷积核进行卷积运算与叠加,提高中心卷积核的决策作用,在推理阶段将脉冲非对称卷积层和批量归一化层进行合并,实现简化运算。结果表明,基于脉冲非对称卷积算法的图像与事件分类模型在DVS Gesture数据集上分类精度可达98.1%,同时不引入额外的计算量和能耗。

基于不同神经网络模型预测体测成绩

为更加准确地反应当前学生的身体素质情况,基于多个不同的神经网络模型构建体测成绩预测模型,为降低体测成绩中各项目数据之间的相关性,使用主成分分析法对数据集进行处理。使用BP神经网络结合脉冲神经网络构建一个预测模型,加强模型处理分析数据集的能力,提高预测的准确性。在长短期记忆神经网络中加入注意力机制构建另一个预测模型,使模型更加关注数据集中的关键信息。通过实验,预测模型输出的预测值与实际值的重合率高达90%以上,预测准确率整体在95%以上。

自校准首脉冲时间编码神经元模型

由于脉冲神经元具有复杂的时空动力过程且脉冲信息不可导,脉冲神经网络(SNN)的训练一直是一个难题。基于人工神经网络(ANN)转SNN间接训练深度SNN的方法,避免了直接训练深度SNN的难题,但该方法所获得的SNN的性能在很大程度上会受到脉冲信息编码机制的影响。在众多编码机制中,首脉冲时间编码(TTFS)具有良好的生物学基础和更高的能效,但现有TTFS编码采用单脉冲形式,信息表征能力较弱,编码所需时间窗较大。为此,在TTFS的单脉冲编码基础上,增加一个校准脉冲,形成一种自校准首脉冲时间(SC-TTFS)编码机制,并构建相应的SC-TTFS神经元模型。在SC-TTFS中,首脉冲为必定发放的脉冲,而校准脉冲根据首脉冲发放后剩余的膜电位来确定是否发放,用于对编码脉冲所引起的转换量化误差和截断误差进行补偿,同时缩小编码所需的时间窗。通过对多种编码对应的转换误差进行对比分析,以及在多种网络结构上进行ANN-SNN转换实验,验证了所提方法的优越性。采用CIFAR10和CIFAR100数据集,基于VGG和ResNet两种网络结构进行了实验验证。结果表明,所提方法在两类网络结构和两种数据集上均实现了精度无损的ANN-SNN转换,且相较于最先进的同类方法,所提方法所构建的SNN具有最短的网络推理延迟。另外,在VGG结构上,所提方法相比TTFS编码能源效率提升了约80%。

基于软阈值降噪的脉冲卷积神经网络轴承故障诊断方法

针对工业场景下滚动轴承信号易受噪声干扰,导致故障诊断准确率低和稳定性差的问题,本文提出一种基于软阈值降噪的脉冲卷积神经网络诊断方法。该方法使用软阈值滤波去噪,运用带时间标签的卷积层处理二维信号,增强动态特征提取能力。同时,通过引入IF和LIF神经元实现对时域和频域信息的联合编码,并采用替代梯度法进行端到端训练。实验结果显示,在信噪比为6dB时,所提方法的诊断准确率达100%,在信噪比为-6dB时诊断准确率达77.33%,优于其他常用方法,表明所提方法在噪声下具有良好的诊断效果和稳定性。

用于双阈值脉冲神经网络的改进自适应阈值算法

脉冲神经网络(spiking neural network, SNN)由于在神经形态芯片上低功耗和高速计算的独特性质而受到广泛的关注。深度神经网络(deep neural network, DNN)到SNN的转换方法是有效的脉冲神经网络训练方法之一,然而从DNN到SNN的转换过程中存在近似误差,转换后的SNN在短时间步长下遭受严重的性能退化。通过对转换过程中的误差进行详细分析,将其分解为量化和裁剪误差以及不均匀误差,提出了一种改进SNN阈值平衡的自适应阈值算法。通过使用最小化均方误差(MMSE)更好地平衡量化误差和裁剪误差;此外,基于IF神经元模型引入了双阈值记忆机制,有效解决了不均匀误差。实验结果表明,改进算法在CIFAR-10、CIFAR-100数据集以及MIT-BIH心律失常数据库上取得了很好的性能,对于CIFAR10数据集,仅用16个时间步长就实现了93.22%的高精度,验证了算法的有效性。