Artificial Spiking Neural Networks Research Articles

A General DNA-Like Hybrid Symbiosis Framework: An EEG Cognitive Recognition Method.

In electroencephalogram (EEG) cognitive recognition research, the combined use of artificial neural networks (ANNs) and spiking neural networks (SNNs) plays an important role to realize different categories of recognition tasks. However, most of the existing studies focus on the unidirectional interaction between an ANN and a SNN, which may be overly dependent on the performance of ANNs or SNNs. Inspired by the symbiosis phenomenon in nature, in this study, we propose a general DNA-like Hybrid Symbiosis (DNA-HS) framework, which enables mutual learning between the ANN and the SNN generated by this ANN through parametric genetic algorithm and bidirectional interaction mechanism to enhance the optimization ability of the model parameters, resulting in a significant improvement of the performance of the DNA-HS framework in all aspects. By comparing with seven typical EEG cognitive recognition models, the performance of the seven hybrid network frameworks constructed using this method on different EEG-based cognitive recognition tasks are all improved to different degrees, verifying the effectiveness of the proposed method. This unified hybrid network framework similar to the DNA structure is expected to open up a new approach and form a new research paradigm for EEG-based cognitive recognition task.

Spiking representation learning for associative memories.

Networks of interconnected neurons communicating through spiking signals offer the bedrock of neural computations. Our brain's spiking neural networks have the computational capacity to achieve complex pattern recognition and cognitive functions effortlessly. However, solving real-world problems with artificial spiking neural networks (SNNs) has proved to be difficult for a variety of reasons. Crucially, scaling SNNs to large networks and processing large-scale real-world datasets have been challenging, especially when compared to their non-spiking deep learning counterparts. The critical operation that is needed of SNNs is the ability to learn distributed representations from data and use these representations for perceptual, cognitive and memory operations. In this work, we introduce a novel SNN that performs unsupervised representation learning and associative memory operations leveraging Hebbian synaptic and activity-dependent structural plasticity coupled with neuron-units modelled as Poisson spike generators with sparse firing (~1 Hz mean and ~100 Hz maximum firing rate). Crucially, the architecture of our model derives from the neocortical columnar organization and combines feedforward projections for learning hidden representations and recurrent projections for forming associative memories. We evaluated the model on properties relevant for attractor-based associative memories such as pattern completion, perceptual rivalry, distortion resistance, and prototype extraction.

Electrocardiography Classification with Leaky Integrate-and-Fire Neurons in an Artificial Neural Network-Inspired Spiking Neural Network Framework.

Monitoring heart conditions through electrocardiography (ECG) has been the cornerstone of identifying cardiac irregularities. Cardiologists often rely on a detailed analysis of ECG recordings to pinpoint deviations that are indicative of heart anomalies. This traditional method, while effective, demands significant expertise and is susceptible to inaccuracies due to its manual nature. In the realm of computational analysis, Artificial Neural Networks (ANNs) have gained prominence across various domains, which can be attributed to their superior analytical capabilities. Conversely, Spiking Neural Networks (SNNs), which mimic the neural activity of the brain more closely through impulse-based processing, have not seen widespread adoption. The challenge lies primarily in the complexity of their training methodologies. Despite this, SNNs offer a promising avenue for energy-efficient computational models capable of displaying a high-level performance. This paper introduces an innovative approach employing SNNs augmented with an attention mechanism to enhance feature recognition in ECG signals. By leveraging the inherent efficiency of SNNs, coupled with the precision of attention modules, this model aims to refine the analysis of cardiac signals. The novel aspect of our methodology involves adapting the learned parameters from ANNs to SNNs using leaky integrate-and-fire (LIF) neurons. This transfer learning strategy not only capitalizes on the strengths of both neural network models but also addresses the training challenges associated with SNNs. The proposed method is evaluated through extensive experiments on two publicly available benchmark ECG datasets. The results show that our model achieves an overall accuracy of 93.8% on the MIT-BIH Arrhythmia dataset and 85.8% on the 2017 PhysioNet Challenge dataset. This advancement underscores the potential of SNNs in the field of medical diagnostics, offering a path towards more accurate, efficient, and less resource-intensive analyses of heart diseases.

EPHA: An Energy-efficient Parallel Hybrid Architecture for ANNs and SNNs

Artificial neural networks (ANNs) and spiking neural networks (SNNs) are two general approaches to achieve artificial intelligence (AI). The former have been widely used in academia and industry fields; the latter, SNNs, are more similar to biological neural networks and can realize ultra-low power consumption, thus have received widespread research attention. However, due to their fundamental differences in computation formula and information coding, the two methods often require different and incompatible platforms. Alongside the development of AI, a general platform that can support both ANNs and SNNs is necessary. Moreover, there are some similarities between ANNs and SNNs, which leaves room to deploy different networks on the same architecture. However, there is little related research on this topic. Accordingly, this article presents an energy-efficient, scalable, and non-Von Neumann architecture (EPHA) for ANNs and SNNs. Our study combines device-, circuit-, architecture-, and algorithm-level innovations to achieve a parallel architecture with ultra-low power consumption. We use the compensated ferrimagnet to act as both synapses and neurons to store weights and perform dot-product operations, respectively. Moreover, we propose a novel computing flow to reduce the operations across multiple crossbar arrays, which enables our design to conduct large and complex tasks. On a suite of ANN and SNN workloads, the EPHA is 1.6× more power-efficient than a state-of-the-art design, NEBULA, in the ANN mode. In the SNN mode, our design is 4 orders of magnitude more than the Loihi in power efficiency.

Developmental Plasticity-inspired Adaptive Pruning for Deep Spiking and Artificial Neural Networks.

Developmental plasticity plays a prominent role in shaping the brain's structure during ongoing learning in response to dynamically changing environments. However, the existing network compression methods for deep artificial neural networks (ANNs) and spiking neural networks (SNNs) draw little inspiration from brain's developmental plasticity mechanisms, thus limiting their ability to learn efficiently, rapidly, and accurately. This paper proposed a developmental plasticity-inspired adaptive pruning (DPAP) method, with inspiration from the adaptive developmental pruning of dendritic spines, synapses, and neurons according to the "use it or lose it, gradually decay" principle. The proposed DPAP model considers multiple biologically realistic mechanisms (such as dendritic spine dynamic plasticity, activitydependent neural spiking trace, and local synaptic plasticity), with additional adaptive pruning strategy, so that the network structure can be dynamically optimized during learning without any pre-training and retraining. Extensive comparative experiments show consistent and remarkable performance and speed boost with the extremely compressed networks on a diverse set of benchmark tasks for deep ANNs and SNNs, especially the spatio-temporal joint pruning of SNNs in neuromorphic datasets. This work explores how developmental plasticity enables complex deep networks to gradually evolve into brain-like efficient and compact structures, eventually achieving state-of-the-art (SOTA) performance for biologically realistic SNNs.

Extremely energy-efficient, magnetic field-free, skyrmion-based memristors for neuromorphic computing

The human brain can process information more efficiently than computers due to the dynamics of neurons and synapses. Mimicking such a system can lead to the practical implementation of artificial spiking neural networks. Spintronic devices have been shown to be an ideal solution for realizing the hardware required for neuromorphic computing. Skyrmions prove to be an effective candidate as information carriers owing to their topological protection and particle-like nature. Ferrimagnet and antiferromagnet-based spintronics have been employed previously to obtain an ultrafast simulation of artificial synapses and neurons. Here, we have proposed a ferromagnetic device of stack Ta3nmPt3nmCu0.65nmCo0.5nmPt1nm that is capable of ultrafast simulation of artificial neurons and synapses, owing to the high velocity of the stabilized skyrmions in the system. Electrical pulses of nanosecond pulse width were used to control the accumulation and dissipation of skyrmions in the system, analogous to the variations in the synaptic weights. Lateral structure inversion asymmetry is used to bring about a field-free switching in the system, leading to an energy-efficient switching process. Magnetic field-free deterministic switching and low pulse width current pulses drastically reduce energy consumption by 106 times compared to the existing ferromagnet-based neuromorphic devices. Artificial neuron, synapse, and memristor functionalities have been reproduced on the same device with characteristic time scales and field-free switching, better than any existing ferromagnet-based neuromorphic devices. The results recognize ferromagnet-based skyrmions as viable candidates for ultrafast neuromorphic spintronics capable of executing cognitive tasks with extremely high efficiency.

Mental stress recognition on the fly using neuroplasticity spiking neural networks

Mental stress is found to be strongly connected with human cognition and wellbeing. As the complexities of human life increase, the effects of mental stress have impacted human health and cognitive performance across the globe. This highlights the need for effective non-invasive stress detection methods. In this work, we introduce a novel, artificial spiking neural network model called Online Neuroplasticity Spiking Neural Network (O-NSNN) that utilizes a repertoire of learning concepts inspired by the brain to classify mental stress using Electroencephalogram (EEG) data. These models are personalized and tested on EEG data recorded during sessions in which participants listen to different types of audio comments designed to induce acute stress. Our O-NSNN models learn on the fly producing an average accuracy of 90.76% (σ = 2.09) when classifying EEG signals of brain states associated with these audio comments. The brain-inspired nature of the individual models makes them robust and efficient and has the potential to be integrated into wearable technology. Furthermore, this article presents an exploratory analysis of trained O-NSNNs to discover links between perceived and acute mental stress. The O-NSNN algorithm proved to be better for personalized stress recognition in terms of accuracy, efficiency, and model interpretability.

Neuromorphic Systems: Devices, Architecture, and Algorithms

The application of the structure and principles of the human brain opens up great opportunities for creating artificial systems based on silicon technology. The energy efficiency and performance of a biosimilar architecture can be significantly higher compared to the traditional von Neumann architecture. This paper presents an overview of the most promising artificial neural network (ANN) and spiking neural network (SNN) architectures for biosimilar systems, called neuromorphic systems. Devices for biosimilar systems, such as memristors and ferroelectric transistors, are considered for use as artificial synapses that determine the possibility of creating various architectures of neuromorphic systems; methods and rules for training structures to work correctly when mimicking biological learning rules, such as long-term synaptic plasticity. Problems hindering the implementation of biosimilar systems and examples of architectures that have been practically implemented are discussed.

Comprehensive SNN Compression Using ADMM Optimization and Activity Regularization.

As well known, the huge memory and compute costs of both artificial neural networks (ANNs) and spiking neural networks (SNNs) greatly hinder their deployment on edge devices with high efficiency. Model compression has been proposed as a promising technique to improve the running efficiency via parameter and operation reduction, whereas this technique is mainly practiced in ANNs rather than SNNs. It is interesting to answer how much an SNN model can be compressed without compromising its functionality, where two challenges should be addressed: 1) the accuracy of SNNs is usually sensitive to model compression, which requires an accurate compression methodology and 2) the computation of SNNs is event-driven rather than static, which produces an extra compression dimension on dynamic spikes. To this end, we realize a comprehensive SNN compression through three steps. First, we formulate the connection pruning and weight quantization as a constrained optimization problem. Second, we combine spatiotemporal backpropagation (STBP) and alternating direction method of multipliers (ADMMs) to solve the problem with minimum accuracy loss. Third, we further propose activity regularization to reduce the spike events for fewer active operations. These methods can be applied in either a single way for moderate compression or a joint way for aggressive compression. We define several quantitative metrics to evaluate the compression performance for SNNs. Our methodology is validated in pattern recognition tasks over MNIST, N-MNIST, CIFAR10, and CIFAR100 datasets, where extensive comparisons, analyses, and insights are provided. To the best of our knowledge, this is the first work that studies SNN compression in a comprehensive manner by exploiting all compressible components and achieves better results.

Neuromorphic brain interfacing and the challenge of human subjectivation

Bidirectional neuromorphic brain interfaces, based on interconnecting brain networks with artificial spiking neural networks, aim to optimize neural prostheses to help people with paralysis to regain autonomy. Their emergence questions the concept of human subjectivation, the continuous process by which we become and remain the subject of our life.

Memory-Dependent Computation and Learning in Spiking Neural Networks Through Hebbian Plasticity.

Spiking neural networks (SNNs) are the basis for many energy-efficient neuromorphic hardware systems. While there has been substantial progress in SNN research, artificial SNNs still lack many capabilities of their biological counterparts. In biological neural systems, memory is a key component that enables the retention of information over a huge range of temporal scales, ranging from hundreds of milliseconds up to years. While Hebbian plasticity is believed to play a pivotal role in biological memory, it has so far been analyzed mostly in the context of pattern completion and unsupervised learning in artificial and SNNs. Here, we propose that Hebbian plasticity is fundamental for computations in biological and artificial spiking neural systems. We introduce a novel memory-augmented SNN architecture that is enriched by Hebbian synaptic plasticity. We show that Hebbian enrichment renders SNNs surprisingly versatile in terms of their computational as well as learning capabilities. It improves their abilities for out-of-distribution generalization, one-shot learning, cross-modal generative association, language processing, and reward-based learning. This suggests that powerful cognitive neuromorphic systems can be built based on this principle.

Design of all-phase-change-memory spiking neural network enabled by Ge-Ga-Sb compound

The implementation of artificial spiking neural network (SNN) usually takes advantage of multiple heterogeneous circuits to mimic either neurons which generate spiking pulses, or synapses which store the weights of event correlations. Here, we design a homogeneous device using Ge-Ga-Sb (GGS) as a phase-change-memory (PCM) material which can do both jobs. The GGS compound shows high stability when used in data storage, such as high working temperature (281°C) and high 10-years data retention temperature (230°C), as well as low resistance drift. Interestingly, when the as-fabricated GGS device is set by iterative narrow-width electric pulses, it first experiences an abrupt resistance drop by two orders of magnitude, followed by a continuous resistance decrease. This unique abrupt-to-progressive transition can be used to mimic both neuronal and synaptic functions, mechanistically enabled by the formation of conductive channels and the continuous growth with the phase separation of crystalline areas. To this end, we propose an all-PCM SNN, which is emulated to have high accuracy (90%) in the standard pattern recognition.

A model of TaOx threshold switching memristor for neuromorphic computing

In this article, we analyzed the experimental data based on the TaOx memristor and found that the threshold switching (TS) characteristics are related to temperature, and its logarithmic I–V curve is in good agreement with the space charge limiting current conduction mechanism. We use this mechanism to establish a TS physical model and then use the physical model to build an LTspice model. The model data are fitted with the experimental data, which is basically consistent. Next, using the TS memristor to simulate a leaky integrate-and-fire neuron circuit, the basic dynamics are realized. By changing the external temperature of the memristor, the output frequency of the neuron will be more intense as the temperature increases. Finally, an artificial spiking neural network (SNN) was built based on this neuron circuit for MNIST recognition task. In this SNN, the input signals fused both voltage amplitude and temperature to achieve neuromorphic multimodal preprocessing and enhance the recognition accuracy. These results demonstrated the reliability of the model, which enhanced the flexibility for exploring the application of TaOx-based TS memristors.

BCImat: a Matlab-based framework for Intracortical Brain-Computer Interfaces and their simulation with an artificial spiking neural network

Ferrea et al., (2022). BCImat: a Matlab-based framework for Intracortical Brain-Computer Interfaces and their simulation with an artificial spiking neural network. Journal of Open Source Software, 7(75), 3956, https://doi.org/10.21105/joss.03956

Realization of volatile and non-volatile resistive switching with N-TiO2 nanorod arrays based memristive devices through compositional control

It is attractive to manipulate the volatile and non-volatile resistive switching behaviors of memristors to create genuine neuromorphic systems such as artificial neural networks (ANNs) and spiking neural networks (SNNs). To investigate the above behaviors, nitrogen has been introduced into TiO2 nanorod arrays (TiO2 NRAs) by hydrothermal processing. X-ray photoelectron spectroscopy (XPS) analysis shows that both lattice and interstitial nitrogen are incorporated into the TiO2 NRAs. Nitrogen doped TiO2 NRAs (N-TiO2 NRAs) based memristive devices with different contents of lattice nitrogen were investigated systematically. The results show that non-volatile restive switching is achieved at up to a relative lattice nitrogen content of 21%, with volatile switching behavior above this value. The volatile memristive devices show similar spiking and decay features to the Leaky Integrate-and-Fire (LIF) model. The lifetime of the decay process of the volatile devices rang from 0.029 to 1.835 s. The longest lifetime is obtained when the relative content of lattice nitrogen is about 70%. The volatile behavior is related to the large defect concentration, caused by nitrogen doping. The proposed nitrogen engineered memristive devices could pave the way to achieving a physical neuromorphic system.

SpikeConverter: An Efficient Conversion Framework Zipping the Gap between Artificial Neural Networks and Spiking Neural Networks

Spiking Neural Networks (SNNs) have recently attracted enormous research interest since their event-driven and brain-inspired structure enables low-power computation. In image recognition tasks, the best results are achieved by SNN so far utilizing ANN-SNN conversion methods that replace activation functions in artificial neural networks~(ANNs) with integrate-and-fire neurons. Compared to source ANNs, converted SNNs usually suffer from accuracy loss and require a considerable number of time steps to achieve competitive accuracy. We find that the performance degradation of converted SNN stems from the fact that the information capacity of spike trains in transferred networks is smaller than that of activation values in source ANN, resulting in less information being passed during SNN inference. To better correlate ANN and SNN for better performance, we propose a conversion framework to mitigate the gap between the activation value of source ANN and the generated spike train of target SNN. The conversion framework originates from exploring an identical relation in the conversion and exploits temporal separation scheme and novel neuron model for the relation to hold. We demonstrate almost lossless ANN-SNN conversion using SpikeConverter for VGG-16, ResNet-20/34, and MobileNet-v2 SNNs on challenging datasets including CIFAR-10, CIFAR-100, and ImageNet. Our results also show that SpikeConverter achieves the abovementioned accuracy across different network architectures and datasets using 32X - 512X fewer inference time-steps than state-of-the-art ANN-SNN conversion methods.

Machine learning with a snapshot of data: Spiking neural network 'predicts' reinforcement histories of pigeons' choice behavior.

An accumulated body of choice research has demonstrated that choice behavior can be understood within the context of its history of reinforcement by measuring response patterns. Traditionally, work on predicting choice behaviors has been based on the relationship between the history of reinforcement—the reinforcer arrangement used in training conditions—and choice behavior. We suggest an alternative method that treats the reinforcement history as unknown and focuses only on operant choices to accurately predict (more precisely, retrodict) reinforcement histories. We trained machine learning models known as artificial spiking neural networks (SNNs) on previously published pigeon datasets to detect patterns in choices with specific reinforcement histories—seven arranged concurrent variable‐interval schedules in effect for nine reinforcers. Notably, SNN extracted information from a small ‘window’ of observational data to predict reinforcer arrangements. The models' generalization ability was then tested with new choices of the same pigeons to predict the type of schedule used in training. We examined whether the amount of the data provided affected the prediction accuracy and our results demonstrated that choices made by the pigeons immediately after the delivery of reinforcers provided sufficient information for the model to determine the reinforcement history. These results support the idea that SNNs can process small sets of behavioral data for pattern detection, when the reinforcement history is unknown. This novel approach can influence our decisions to determine appropriate interventions; it can be a valuable addition to our toolbox, for both therapy design and research.

Using complex behavior to understand brain mechanisms in health and disease.

Using complex behavior to understand brain mechanisms in health and disease.

Progressive compression and weight reinforcement for spiking neural networks

SummaryNeuromorphic architectures are one of the most promising architectures to reduce the energy consumption of tomorrow's computers. These architectures are inspired by the behavior of the brain at a fairly precise level and consist of artificial spiking neural networks. To optimize the implementation of these architectures, we propose in this article a novel progressive network compression and reinforcement technique. This technique consists of two functions: progressive pruning and dynamic synaptic weight reinforcement, which we apply after each training batch. The proposed approach delivers a highly compressed network (up to 80% of compression rate) while preserving the network performance when tested with MNIST.

Arrays of Nanocomposite Crossbar Memristors for the Implementation of Formal and Spiking Neuromorphic Systems

Memristors are promising devices for use as synapses in hardware-based artificial formal and spiking neural networks. Their use can increase the energy efficiency of such networks when solving various cognitive tasks. At the same time, for the implementation of hardware-based neural networks, a large number of rather densely packed structures are required, which can be achieved by creating memristive matrices in a crossbar architecture. Crossbar arrays of nanocomposite memristors on silicon substrates with protection against edge effects are developed and manufactured, their resistive switching is investigated, a formal neural network is created on their basis, which is capable of recognizing simple patterns after loading a pretrained weight map, and the possibility of precise adjustment of the conductivity of the crossbar memristors by various means applicable in the creation of more complex spiking and formal neural networks is demonstrated.