Search

In the field of neuroscience, a significant challenge lies in extracting essential features from biological signals like Electroencephalography (EEG). Utilized as a non-invasive method, EEG records brain activities through metal electrodes on the scalp. The analysis of EEG data finds applications in various domains, including concealed information tests, aimed at detecting deception. This paper introduces the Spatial Spiking Neural Network, a supervised approach for classifying EEG data collected during concealed information tests. Temporal EEG data undergoes filtration using a Finite Impulse Response (FIR) filter, while Common Spatial Pattern (CSP) is employed to extract spatial components. Binary classification is achieved through an integrate-and-fire neuron model, where the frequency of spike generation determines the classification. Spiking Neural Networks (SNNs) offers advantages in terms of temporal precision, event-driven processing, and low power consumption. Their spike-based communication allows for efficient handling of sparse data and recognition of temporal patterns, contributing to robustness and energy efficiency. The proposed model is applied separately to each subject's EEG data, and the results are compared with traditional classification algorithms. The proposed approach attains a peak accuracy of 90.15%, showcasing superior performance compared to alternative methods. [ABSTRACT FROM AUTHOR]

Understanding cognitive processes in the brain demands sophisticated models capable of replicating neural dynamics at large scales. We present a physiologically inspired speech recognition architecture, compatible and scalable with deep learning frameworks, and demonstrate that end-to-end gradient descent training leads to the emergence of neural oscillations in the central spiking neural network. Significant cross-frequency couplings, indicative of these oscillations, are measured within and across network layers during speech processing, whereas no such interactions are observed when handling background noise inputs. Furthermore, our findings highlight the crucial inhibitory role of feedback mechanisms, such as spike frequency adaptation and recurrent connections, in regulating and synchronizing neural activity to improve recognition performance. Overall, on top of developing our understanding of synchronization phenomena notably observed in the human auditory pathway, our architecture exhibits dynamic and effcient information processing, with relevance to neuromorphic technology. [ABSTRACT FROM AUTHOR]

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 (~1Hz mean and ~100 Hz maximum firing rate). Crucially, the architecture of our model derives from the neocortical columnar organization and combines feed forward 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. [ABSTRACT FROM AUTHOR]

Acoustic perception of the automotive environment has the potential to advance driving potentials with enhanced safety. The challenge arises when these acoustic perception systems need to perform under resource and power constraints on edge devices. Neuromorphic computing has introduced spiking neural networks in the context of ultra-low power sensory edge devices. Spiking architectures leverage biological plausibility to achieve computational capabilities, accurate performance, and great compatibility with neuromorphic hardware. In this work, we explore the depths of spiking neurons and feature components with the acoustic scene analysis task for siren sounds. This research work aims to address the qualitative analysis of sliding windows' variation on the feature extraction front of the preprocessing pipeline. Optimization of the parameters to exploit the feature extraction stage facilitates the advancement of the performance of the acoustics anomaly detection task. We exploit the parameters for mel spectrogram features and FFT calculations, prone to be suitable for computations in hardware. We conduct experiments with different window sizes and the overlapping ratio within the windows. We present our results for performance measures like accuracy and onset latency to provide an insight on the choice of optimal window. The non-trivial motivation of this research is to understand the effect of encoding behavior of spiking neurons with different windows. We further investigate the heterogeneous nature of membrane and synaptic time constants and their impact on the accuracy of anomaly detection. On a large scale audio dataset comprising of siren sounds and road traffic noises, we obtain accurate predictions of siren sounds using a recurrent spiking neural network. The baseline dataset comprising siren and noise sequences is enriched with a bird dataset to evaluate the model with unseen samples. [ABSTRACT FROM AUTHOR]

Spike timing-dependent plasticity (STDP) is crucial for training neural networks (SNNs), offering a hardware-compatible and energy-efficient alternative to backpropagation. Current STDP hardware platforms encounter significant challenges, such as slowness, high energy consumption, and limited configurability. To overcome these issues, this paper presents a high-performance SNN training platform. A parallel multi-ring first-in-first-out structure with event-driven processing for spike handling is proposed, which enhances training efficiency, and the flexible pre- and post-synaptic parallelism enhances speed and flexibility. A dataflow strategy that considers spike sparsity unifies spike representations by updating weights only upon spike arrival, thereby promoting logical symmetry and enabling parallelization. Additionally, three encoding strategies, including a hybrid encoding, are implemented to address diverse scenarios. Leveraging Xilinx field-programmable gate array and Jetson Xavier NX, the proposed platform achieved remarkable performance gains. On the MNIST dataset, the platform demonstrated a 22.51 × speedup, 2.13% accuracy boost, and 14.79 × reduction in energy consumption. On the Fashion-MNIST dataset, it improved accuracy by 10.74% and the training speed by 1.89 × . Thus, this platform can significantly advance the SNN training efficiency and performance. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) which use spiking neurons as a component, have shown substantial promise in simulating biological neuron mechanisms and saving computing power. However, preset or suboptimal hyperparameters are still used for spiking neurons adopted in SNNs, and the heterogeneity of neurons is limited, limiting SNNs inference accuracy. Inspired by neuroscience observations that hyperparameters are related to the membrane potential in neurons, in this paper, a new module for implementing adaptive hyperparameters dynamically is proposed, enabling flexible hyperparameters to be obtained for spiking neurons. In addition, inspired by neuroscience observations that heterogeneity in current drive force of synaptic integration process, we propose a new module to distribute synaptic driving force factors in neurons to maximize synaptic integration rationalization. Utilizing these methods enables SNNs to have fast convergence capability and appropriate inference of neuron dynamics. Finally, the effects of the proposed adaptive hyperparameters and driving force distribution mechanism are evaluated on different datasets. The results show that SNNs with our methods have improved accuracy on all test datasets, exhibit robustness to different initial hyperparameters, and exhibit more realistic biological behavior. [ABSTRACT FROM AUTHOR]

In the era of large-scale pretrained models, artificial neural networks (ANNs) have excelled in natural language understanding (NLU) tasks. However, their success often necessitates substantial computational resources and energy consumption. To address this, we explore the potential of spiking neural networks (SNNs) in NLU—a promising avenue with demonstrated advantages, including reduced power consumption and improved efficiency due to their event-driven characteristics. We propose the SpikingMiniLM, a novel spiking Transformer model tailored for natural language understanding. We first introduce a multistep encoding method to convert text embeddings into spike trains. Subsequently, we redesign the attention mechanism and residual connections to make our model operate on the pure spike-based paradigm without any normalization technique. To facilitate stable and fast convergence, we propose a general parameter initialization method grounded in the stable firing rate principle. Furthermore, we apply an ANN-to-SNN knowledge distillation to overcome the challenges of pretraining SNNs. Our approach achieves a macro-average score of 75.5 on the dev sets of the GLUE benchmark, retaining 98% of the performance exhibited by the teacher model MiniLMv2. Our smaller model also achieves similar performance to BERTMINI with fewer parameters and much lower energy consumption, underscoring its competitiveness and resource efficiency in NLU tasks. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) take inspiration from the brain to enable energy-efficient computations. Since the advent of Transformers, SNNs have struggled to compete with artificial networks on modern sequential tasks, as they inherit limitations from recurrent neural networks (RNNs), with the added challenge of training with non-differentiable binary spiking activations. However, a recent renewed interest in efficient alternatives to Transformers has given rise to state-of-the-art recurrent architectures named state space models (SSMs). This work systematically investigates, for the first time, the intersection of state-of-the-art SSMs with SNNs for long-range sequence modelling. Results suggest that SSM-based SNNs can outperform the Transformer on all tasks of a well-established long-range sequence modelling benchmark. It is also shown that SSM-based SNNs can outperform current state-of-the-art SNNs with fewer parameters on sequential image classification. Finally, a novel feature mixing layer is introduced, improving SNN accuracy while challenging assumptions about the role of binary activations in SNNs. This work paves the way for deploying powerful SSM-based architectures, such as large language models, to neuromorphic hardware for energy-efficient long-range sequence modelling. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) have received increasing attention due to their high biological plausibility and energy efficiency. The binary spike-based information propagation enables efficient sparse computation in event-based and static computer vision applications. However, the weight precision and especially the membrane potential precision remain as high-precision values (e.g., 32 bits) in state-of-the-art SNN algorithms. Each neuron in an SNN stores the membrane potential over time and typically updates its value in every time step. Such frequent read/write operations of high-precision membrane potential incur storage andmemory access overhead in SNNs, which undermines the SNNs' compatibility with resource-constrained hardware. To resolve this inefficiency, prior works have explored the time step reduction and low-precision representation of membrane potential at a limited scale and reported significant accuracy drops. Furthermore, while recent advances in on-device AI present pruning and quantization optimization with different architectures and datasets, simultaneous pruning with quantization is highly under-explored in SNNs. In this work, we present SpQuant-SNN, a fully-quantized spiking neural network with ultra-low precision weights, membrane potential, and high spatial-channel sparsity, enabling the end-to-end low precision with significantly reduced operations on SNN. First, we propose an integer-only quantization scheme for the membrane potential with a stacked surrogate gradient function, a simple-yet-effective method that enables the smooth learning process of quantized SNN training. Second, we implement spatial-channel pruning with membrane potential prior, toward reducing the layer-wise computational complexity, and floating-point operations (FLOPs) in SNNs. Finally, to further improve the accuracy of low-precision and sparse SNN, we propose a self-adaptive learnable potential threshold for SNN training. Equipped with high biological adaptiveness, minimal computations, and memory utilization, SpQuant-SNN achieves state-of-the-art performance across multiple SNN models for both event-based and static image datasets, including both image classification and object detection tasks. The proposed SpQuant-SNN achieved up to 13× memory reduction and >4.7× FLOPs reduction with <1.8% accuracy degradation for both classification and object detection tasks, compared to the SOTA baseline. [ABSTRACT FROM AUTHOR]

In remote sensing scene classification (RSSC), restrictions on real-time processing on power consumption, performance, and resources necessitate the compression of neural networks. Unlike artificial neural networks (ANNs), spiking neural networks (SNNs) convey information through spikes, offering superior energy efficiency and biological plausibility. However, the high latency of SNNs restricts their practical application in RSSC. Therefore, there is an urgent need to research ultra-low-latency SNNs. As latency decreases, the performance of the SNN significantly deteriorates. To address this challenge, we propose a novel spatio-temporal pruning method that enhances the feature capture capability of ultra-low-latency SNNs. Our approach integrates spatial fundamental structures during the training process, which are subsequently pruned. We conduct a comprehensive evaluation of the impacts of these structures across classic network architectures, such as VGG and ResNet, demonstrating the generalizability of our method. Furthermore, we develop an ultra-low-latency training framework for SNNs to validate the effectiveness of our approach. In this paper, we successfully achieve high-performance ultra-low-latency SNNs with a single time step for the first time in RSSC. Remarkably, our SNN with one time step achieves at least 200 times faster inference time while maintaining a performance comparable to those of other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Automated handwritten digit identification has become essential for many daily operations such as verifying the authenticity of a cheque or deciphering a postal code. Researchers have been putting a lot of effort into figuring out how to automatically recognize and sort handwritten digits because of the advent of robotic technology in recent decades. Due to the complexity of the Arabic language and a lack of publicly available Arabic handwritten digit datasets, previous research has primarily concentrated on automating the recognition of English and European digits on availability of relevant datasets. Arabic handwritten recognition also plays a significant role in electronic-learning (e-learning) systems. It has not taken into account the recognition of handwritten Arabic digits. Arabic handwritten digits vary in size, form, slant, and image noise, which can cause changes in numeral topology. These factors make it difficult to classify and recognize Arabic handwritten digits. In order to overcome these challenges, we implemented a biologically plausible technique to classify Arabic handwritten digits. The objective of the research is to address the unique challenges of style, size, shape, slant variations, and image noise of Arabic handwritten digits in classification and recognition. To categorize Arabic handwritten digits, we employed convolutional spiking neural network (CSNN) and spike neural network (SNN) models. The reason for using SNN and SCNN is that the second-generation neural networks accumulate accolades on various computer vision tasks, i.e., pattern recognition, segmentation, and classification. However, there are certain challenges, i.e., energy inefficiency and computational cost, presented by these networks. The spiking neural network models also address these issues. These models are low-power, high-performance neural network models. We trained the spiking neural network model on Arabic digit datasets using a rate-based, non-spiking algorithm, i.e., backpropagation and soft-LIF, then integrated them with a spiking neural network. We have also used an STDP (spike time-dependent plasticity)-based spiking neural network for Arabic digit recognition. We attenuate the spiking variability in CSNN by adding noise during training, compensating the training errors, and bringing robustness against the spiking variability. We experimented with soft-LIF, a non-spiking version of the LIF neuron, and a variant of STDP rules for Arabic digit classification. Experimental results illustrate that CSNN achieves high performance at a significantly low energy and computational cost. We attained 98.98% and 91.16% recognition rates for the Arabic digit dataBase (ADBase) dataset using the convolutional spiking model and STDP spiking model, respectively. [ABSTRACT FROM AUTHOR]

Parkinsonism is presented as a motor syndrome characterized by rigidity, tremors, and bradykinesia, with Parkinson's disease (PD) being the predominant cause. The discovery that those motor symptoms result from the death of dopaminergic cells in the substantia nigra led to focus most of parkinsonism research on the basal ganglia (BG). However, recent findings point to an active involvement of the cerebellum in this motor syndrome. Here, we have developed a multiscale computational model of the rodent brain's BG–cerebellar network. Simulations showed that a direct effect of dopamine depletion on the cerebellum must be taken into account to reproduce the alterations of neural activity in parkinsonism, particularly the increased beta oscillations widely reported in PD patients. Moreover, dopamine depletion indirectly impacted spike-time-dependent plasticity at the parallel fiber-Purkinje cell synapses, degrading associative motor learning as observed in parkinsonism. Overall, these results suggest a relevant involvement of cerebellum in parkinsonism associative motor symptoms. [ABSTRACT FROM AUTHOR]

Spiking Neural Network (SNN), originating from the neural behavior in biology, has been recognized as one of the next-generation neural networks. Conventionally, SNNs can be obtained by converting from pre-trained Artificial Neural Networks (ANNs) by replacing the non-linear activation with spiking neurons without changing the parameters. In this work, we argue that simply copying and pasting the weights of ANN to SNN inevitably results in activation mismatch, especially for ANNs that are trained with batch normalization (BN) layers. To tackle the activation mismatch issue, we first provide a theoretical analysis by decomposing local layer-wise conversion error, and then quantitatively measure how this error propagates throughout the layers using the second-order analysis. Motivated by the theoretical results, we propose a set of layer-wise parameter calibration algorithms, which adjusts the parameters to minimize the activation mismatch. To further remove the dependency on data, we propose a privacy-preserving conversion regime by distilling synthetic data from source ANN and using it to calibrate the SNN. Extensive experiments for the proposed algorithms are performed on modern architectures and large-scale tasks including ImageNet classification and MS COCO detection. We demonstrate that our method can handle the SNN conversion and effectively preserve high accuracy even in 32 time steps. For example, our calibration algorithms can increase up to 63% accuracy when converting MobileNet against baselines. [ABSTRACT FROM AUTHOR]

Bioinspired approaches tend to mimic some biological functions for the purpose of creating more efficient and robust systems. These can be implemented in both software and hardware designs. A neuromorphic software part can include, for example, Spiking Neural Networks (SNNs) or event-based representations. Regarding the hardware part, we can find different sensory systems, such as Dynamic Vision Sensors, touch sensors, and actuators, which are linked together through specific interface boards. To run real-time SNN models, specialised hardware such as SpiNNaker, Loihi, and TrueNorth have been implemented. However, neuromorphic computing is still in development, and neuromorphic platforms are still not easily accessible to researchers. In addition, for Neuromorphic Robotics, we often need specially designed and fabricated PCBs for communication with peripheral components and sensors. Therefore, we developed an all-in-one neuromorphic system that emulates neuromorphic computing by running a Virtual Machine on a conventional low-power CPU. The Virtual Machine includes Python and Brian2 simulation packages, which allow the running of SNNs, emulating neuromorphic hardware. An additional, significant advantage of using conventional hardware such as Raspberry Pi in comparison to purpose-built neuromorphic hardware is that we can utilise the built-in physical input–output (GPIO) and USB ports to directly communicate with sensors. As a proof of concept platform, a robotic goalkeeper has been implemented, using a Raspberry Pi 5 board and SNN model in Brian2. All the sensors, namely DVS128, with an infrared module as the touch sensor and Futaba S9257 as the actuator, were linked to a Raspberry Pi 5 board. We show that it is possible to simulate SNNs on a conventional low-power CPU running real-time tasks for low-latency and low-power robotic applications. Furthermore, the system excels in the goalkeeper task, achieving an overall accuracy of 84% across various environmental conditions while maintaining a maximum power consumption of 20 W. Additionally, it reaches 88% accuracy in the online controlled setup and 80% in the offline setup, marking an improvement over previous results. This work demonstrates that the combination of a conventional low-power CPU running a Virtual Machine with only selected software is a viable competitor to neuromorphic computing hardware for robotic applications. [ABSTRACT FROM AUTHOR]

The flexibility, adaptability, and resilience of even simple brains are unmatched by any current technology. Recent unexpected difficulties in realising truly autonomous vehicles, making reliable medical diagnoses, detecting offensive online content and even just recognising faces, show that brains remain significantly functionally more capable than we can currently emulate. Fittingly, in recent years we have made significant progress identifying computational principles that underlie neural function. We are beginning to dispense with the overly simplistic stimulus-driven encode/transmit/decode doctrine. Instead we must embrace the brain's inherent dynamic complexity and emergent properties and explain how plasticity moulds the dynamics to capture useful couplings across brain regions and between the brain, the body, and the world. While certainly not complete, we have sufficient evidence that a synthesis of these ideas could result in a deeper understanding of neural computation and which could potentially be used to construct new AI technologies with unique capabilities. I discuss the relevant neuroscientific principles, the advantages they have for computation, and how they can benefit AI. Limitations of current AI are now generally recognised. I postulate that we understand enough about the brain to immediately offer novel AI formulations. [ABSTRACT FROM AUTHOR]

The integration of Spiking Neural Networks (SNNs) into the analysis and interpretation of physiological and speech signals has emerged as a groundbreaking approach, offering enhanced performance and deeper insights into the underlying biological processes. This review aims to summarize key advances, methodologies, and applications of SNNs within these domains, highlighting their unique ability to mimic the temporal dynamics and efficiency of the human brain. We dive into the core principles of SNNs, their neurobiological underpinnings, and the computational advantages they bring to signal processing, particularly in handling the temporal and spatial complexities inherent in physiological and speech data. Comparative analyses with conventional neural network models are presented to underscore the superior efficiency, lower power consumption, and higher temporal resolution of SNNs. The review further explores challenges and future prospects, highlighting the potential of SNNs to revolutionize wearable healthcare monitoring systems, neuroprosthetic devices, and natural language processing technologies. By providing a comprehensive overview of current strategies, this review aims to inspire innovative approaches in the field, fostering advances in real-time and energy-efficient processing of complex biological signals. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has had a significant impact on human life because of its pervasiveness across industries and its rapid development. Although AI has achieved superior performance in learning and reasoning, it encounters challenges such as substantial computational demands, privacy concerns, communication delays, and high energy consumption associated with cloud-based models. These limitations have facilitated a paradigm change in on-device AI processing, which offers enhanced privacy, reduced latency, and improved power efficiency through the direct execution of computations on devices. With advancements in neuromorphic systems, spiking neural networks (SNNs), often referred to as the next generation of AI, are currently in focus as on-device AI. These technologies aim to mimic the human brain efficiency and provide promising real-time processing with minimal energy. This study reviewed the application of SNNs in the analysis of biomedical signals (electroencephalograms, electrocardiograms, and electromyograms), and consequently, investigated the distinctive attributes and prospective future paths of SNNs models in the field of biomedical signal analysis. [ABSTRACT FROM AUTHOR]

As the third‐generation network succeeding artificial neural networks (ANNs), spiking neural networks (SNNs) offer high robustness and low energy consumption. Inspired by biological systems, the limitations of low robustness and high‐power consumption in deep reinforcement learning (DRL) are addressed by introducing SNNs. The Brain Q‐network (BrainQN) is proposed, which replaces the neurons in the classic Deep Q‐learning (DQN) algorithm with SNN neurons. BrainQN is trained using surrogate gradient learning (SGL) and ANN‐to‐SNN conversion methods. Robustness tests with input noise reveal BrainQN's superior performance, achieving an 82.14% increase in rewards under low noise and 71.74% under high noise compared to DQN. These findings highlight BrainQN's robustness and superior performance in noisy environments, supporting its application in complex scenarios. SGL‐trained BrainQN is more robust than ANN‐to‐SNN conversion under high noise. The differences in network output correlations between noisy and original inputs, along with training algorithm distinctions, explain this phenomenon. BrainQN successfully transitioned from a simulated Pong environment to a ball‐catching robot with dynamic vision sensors (DVS). On the neuromorphic chip PAICORE, it shows significant advantages in latency and power consumption compared to Jetson Xavier NX. [ABSTRACT FROM AUTHOR]

Integrating RGB and Event (RGBE) multi-domain information obtained by high-dynamic-range and temporal-resolution event cameras has been considered an effective scheme for robust object tracking. However, existing RGBE tracking methods have overlooked the unique spatio-temporal features over different domains, leading to object tracking failure and ineffeciency, especally for objects against complex backgrounds. To address this problem, we propose a novel tracker based on adaptive-time feature extraction hybrid networks, namely Siamese Event Frame Tracker (SiamEFT), which focuses on the effective representation and utilization of the diverse spatio-temporal features of RGBE. We first design an adaptive-time attention module to aggregate event data into frames based on adaptive-time weights to enhance information representation. Subsequently, the SiamEF module and cross-network fusion module combining artificial neural networks and spiking neural networks hybrid network are designed to effectively extract and fuse the spatio-temporal features of RGBE. Extensive experiments on two RGBE datasets (VisEvent and COESOT) show that the SiamEFT achieves a success rate of 0.456 and 0.574, outperforming the state-of-the-art competing methods and exhibiting a 2.3-fold enhancement in effeciency. These results validate the superior accuracy and effeciency of SiamEFT in diverse and challenging scenes. [ABSTRACT FROM AUTHOR]

In this study, a novel control scheme based on spiking neural networks (SNNs) has been proposed to accomplish the trajectory tracking of quadrotor unmanned aerial vehicles (UAVs). The update rules for the network parameters have been derived using the Lyapunov stability theorem. Three different trajectories have been utilized in the simulated and experimental studies to verify the efficacy of the proposed control scheme. The acquired results have been compared with the responses obtained for proportional–integral–derivative (PID) and traditional neural network controllers. Simulated and experimental studies demonstrate that the proposed SNN-based controller is capable of providing better tracking accuracy and robust system response in the presence of disturbing factors. [ABSTRACT FROM AUTHOR]

A long-standing research goal is to develop computing technologies that mimic the brain's capabilities by implementing computation in electronic systems directly inspired by its structure, function, and operational mechanisms, using low-power, spike-based neural networks. The Loihi neuromorphic processor provides a low-power, large-scale network of programmable silicon neurons for brain-inspired artificial intelligence applications. This paper exploits the Loihi processors and a theory-guided methodology to enable unsupervised learning of spike patterns. Our method ensures efficient and rapid selection of the network's hyperparameters, enabling the neuromorphic processor to generate attractor states through real-time unsupervised learning. Precisely, we follow a fast design process in which we fine-tune network parameters using mean-field theory. Moreover, we measure the network's learning ability regarding its error correction and pattern completion aptitude. Finally, we observe the dynamic energy consumption of the neuron cores for each millisecond of simulation equal to 23 μ J/time step during the learning and recall phase for four attractors composed of 512 excitatory neurons and 256 shared inhibitory neurons. This study showcases how large-scale, low-power digital neuromorphic processors can be quickly programmed to enable the autonomous generation of attractor states. These attractors are fundamental computational primitives that theoretical analysis and experimental evidence indicate as versatile and reusable components suitable for a wide range of cognitive tasks. [ABSTRACT FROM AUTHOR]

Recent trends have shown that autonomous agents, such as Autonomous Ground Vehicles (AGVs), Unmanned Aerial Vehicles (UAVs), and mobile robots, effectively improve human productivity in solving diverse tasks. However, since these agents are typically powered by portable batteries, they require extremely low power/energy consumption to operate in a long lifespan. To solve this challenge, neuromorphic computing has emerged as a promising solution, where bio-inspired Spiking Neural Networks (SNNs) use spikes from event-based cameras or data conversion pre-processing to perform sparse computations efficiently. However, the studies of SNN deployments for autonomous agents are still at an early stage. Hence, the optimization stages for enabling efficient embodied SNN deployments for autonomous agents have not been defined systematically. Toward this, we propose a novel framework called SNN4Agents that consists of a set of optimization techniques for designing energy-efficient embodied SNNs targeting autonomous agent applications. Our SNN4Agents employs weight quantization, timestep reduction, and attention window reduction to jointly improve the energy efficiency, reduce the memory footprint, optimize the processing latency, while maintaining high accuracy. In the evaluation, we investigate use cases of event-based car recognition, and explore the trade-offs among accuracy, latency, memory, and energy consumption. The experimental results show that our proposed framework can maintain high accuracy (i.e., 84.12% accuracy) with 68.75% memory saving, 3.58x speed-up, and 4.03x energy efficiency improvement as compared to the state-of-the-art work for the NCARS dataset. In this manner, our SNN4Agents framework paves the way toward enabling energy-efficient embodied SNN deployments for autonomous agents. [ABSTRACT FROM AUTHOR]

Direct training of Spiking Neural Networks (SNNs) on neuromorphic hardware has the potential to significantly reduce the energy consumption of artificial neural network training. SNNs trained with Spike Timing-Dependent Plasticity (STDP) benefit from gradient-free and unsupervised local learning, which can be easily implemented on ultra-low-power neuromorphic hardware. However, classification tasks cannot be performed solely with unsupervised STDP. In this paper, we propose Stabilized Supervised STDP (S2-STDP), a supervised STDP learning rule to train the classification layer of an SNN equipped with unsupervised STDP for feature extraction. S2-STDP integrates error-modulated weight updates that align neuron spikes with desired timestamps derived from the average firing time within the layer. Then, we introduce a training architecture called Paired Competing Neurons (PCN) to further enhance the learning capabilities of our classification layer trained with S2-STDP. PCN associates each class with paired neurons and encourages neuron specialization toward target or non-target samples through intra-class competition. We evaluate our methods on image recognition datasets, including MNIST, Fashion-MNIST, and CIFAR10. Results show that our methods outperform state-of-the-art supervised STDP learning rules, for comparable architectures and numbers of neurons. Further analysis demonstrates that the use of PCN enhances the performance of S2-STDP, regardless of the hyperparameter set and without introducing any additional hyperparameters. [ABSTRACT FROM AUTHOR]

Inspired by the motion processing pathway, this paper proposes a bio-inspired feedforward spiking network model based on Hodgkin–Huxley neurons for human motion perception. The proposed network mimics the mechanisms of direction selectivity found in simple and complex cells of the primary visual cortex. Simple cells' receptive fields are modeled using Gabor energy filters, while complex cells' receptive fields are constructed by integrating the responses of simple cells in an energy model. To generate the motion map, the spiking output of the network integrates motion information encoded by the responses of complex cells with various preferred directions. Simulation results demonstrate that the spiking neuron-based network effectively replicates the directional selectivity operation of the visual cortex when presented with a sequence of time-varying images. We evaluate the proposed model against state-of-the-art spiking neuron-based motion detection models using publicly available datasets. The results highlight the model's capability to extract motion energy from diverse video sequences, akin to human visual motion perception models. Additionally, we showcase the application of the proposed model in motion segmentation tasks and compare its performance with state-of-the-art motion-based segmentation models using challenging video segmentation benchmarks. The results indicate competitive performance. The motion maps generated by the proposed model can be utilized for action recognition in input videos. [ABSTRACT FROM AUTHOR]

Neuromorphic computing, a brain-inspired non-Von Neumann computing system, addresses the challenges posed by the Moore's law memory wall phenomenon. It has the capability to enhance performance while maintaining power efficiency. Neuromorphic chip architecture requirements vary depending on the application and optimising it for large-scale applications remains a challenge. Neuromorphic chips are programmed using spiking neural networks which provide them with important properties such as parallelism, asynchronism, and on-device learning. Widely used spiking neuron models include the Hodgkin–Huxley Model, Izhikevich model, integrate-and-fire model, and spike response model. Hardware implementation platforms of the chip follow three approaches: analogue, digital, or a combination of both. Each platform can be implemented using various memory topologies which interconnect with the learning mechanism. Current neuromorphic computing systems typically use the unsupervised learning spike timing-dependent plasticity algorithms. However, algorithms such as voltage-dependent synaptic plasticity have the potential to enhance performance. This review summarises the potential neuromorphic chip architecture specifications and highlights which applications they are suitable for. [ABSTRACT FROM AUTHOR]

Brain‐inspired parallel computing is increasingly considered a solution to overcome memory bottlenecks, driven by the surge in data volume. Extensive research has focused on developing memristor arrays, energy‐efficient computing strategies, and varied operational mechanisms for synaptic devices to enable this. However, to realize truly biologically plausible neuromorphic computing, it is essential to consider temporal and spatial aspects of input signals, particularly for systems based on the leaky integrate‐and‐fire model. This review highlights the significance of neuromorphic computing and outlines the fundamental components of hardware‐based neural networks. Traditionally, neuromorphic computing has relied on two‐terminal devices such as artificial synapses. However, these suffer from significant drawbacks, such as current leakage and the lack of a third terminal for precise synaptic weight adjustment. As alternatives, three‐terminal synaptic devices, including memtransistors, ferroelectric, floating‐gate, and charge‐trapped synaptic devices, as well as optoelectronic options, are explored. For an accurate replication of biological neural networks, it is vital to integrate artificial neurons and synapses, implement neurobiological functions in hardware, and develop sensory neuromorphic computing systems. This study delves into the operational mechanisms of these artificial components and discusses the integration process necessary for realizing biologically plausible neuromorphic computing, paving the way for future brain‐inspired electronic systems. [ABSTRACT FROM AUTHOR]

The potential of neuromorphic (NM) solutions often lies in their low-SWaP (Size, Weight, and Power) capabilities, which often drive their application to domains that could benefit from this. Nevertheless, spiking neural networks (SNNs), with their inherent time-based nature, present an attractive alternative also for areas where data features are present in the time dimension, such as time series forecasting. Time series data, characterized by seasonality and trends, can benefit from the unique processing capabilities of SNNs, which offer a novel approach for this type of task. Additionally, time series data can serve as a benchmark for evaluating SNN performance, providing a valuable alternative to traditional datasets. However, the challenge lies in the real-valued nature of time series data, which is not inherently suited for SNN processing. In this work, we propose a novel spike-encoding mechanism and two loss functions to address this challenge. Our encoding system, inspired by NM event-based sensors, converts the derivative of a signal into spikes, enhancing interoperability with the NM technology and also making the data suitable for SNN processing. Our loss functions then optimize the learning of subsequent spikes by the SNN. We train a simple SNN using SLAYER as a learning rule and conduct experiments using two electricity load forecasting datasets. Our results demonstrate that SNNs can effectively learn from encoded data, and our proposed DecodingLoss function consistently outperforms SLAYER's SpikeTime loss function. This underscores the potential of SNNs for time series forecasting and sets the stage for further research in this promising area of research. [ABSTRACT FROM AUTHOR]

Abstract Spiking neural networks (SNNs) take inspiration from the brain to enable energy-efficient computations. Since the advent of Transformers, SNNs have struggled to compete with artificial networks on modern sequential tasks, as they inherit limitations from recurrent neural networks (RNNs), with the added challenge of training with non-differentiable binary spiking activations. However, a recent renewed interest in efficient alternatives to Transformers has given rise to state-of-the-art recurrent architectures named state space models (SSMs). This work systematically investigates, for the first time, the intersection of state-of-the-art SSMs with SNNs for long-range sequence modelling. Results suggest that SSM-based SNNs can outperform the Transformer on all tasks of a well-established long-range sequence modelling benchmark. It is also shown that SSM-based SNNs can outperform current state-of-the-art SNNs with fewer parameters on sequential image classification. Finally, a novel feature mixing layer is introduced, improving SNN accuracy while challenging assumptions about the role of binary activations in SNNs. This work paves the way for deploying powerful SSM-based architectures, such as large language models, to neuromorphic hardware for energy-efficient long-range sequence modelling.

The paper proposes a non-iterative training algorithm for a power efficient SNN classifier for applications in self-learning systems. The approach uses mechanisms of preprocessing of signals from sensory neurons typical of a thalamus in a diencephalon. The algorithm concept is based on a cusp catastrophe model and on training by routing. The algorithm guarantees a zero dispersion of connection weight values across the entire network, which is particularly important in the case of hardware implementation based on programmable logic devices. Due to non-iterative mechanisms inspired by training methods for associative memories, the approach makes it possible to estimate the capacity of the network and required hardware resources. The trained network shows resistance to the phenomenon of catastrophic forgetting. Low complexity of the algorithm makes in-situ hardware training possible without using power-hungry accelerators. The paper compares the complexities of hardware implementations of the algorithm with the classic STDP and conversion procedures. The basic application of the algorithm is an autonomous agent equipped with a vision system and based on a classic FPGA device.

The recent emergence of the hybrid quantum-classical neural network (HQCNN) architecture has garnered considerable attention because of the potential advantages associated with integrating quantum principles to enhance various facets of machine learning algorithms and computations. However, the current investigated serial structure of HQCNN, wherein information sequentially passes from one network to another, often imposes limitations on the trainability and expressivity of the network. In this study, we introduce a novel architecture termed parallel proportional fusion of spiking and quantum neural networks (PPF-SQNN). The dataset information is simultaneously fed into both the spiking neural network and the variational quantum circuits, with the outputs amalgamated in proportion to their individual contributions. We systematically assess the impact of diverse PPF-SQNN parameters on network performance for image classification, aiming to identify the optimal configuration. On three datasets for image classification tasks, the final classification accuracy reached 98.2%, 99.198%, and 97.921%, respectively, with loss values all below 0.2, outperforming the compared serial networks. In noise testing, it also demonstrates good classification performance even under noise intensities of 0.9 Gaussian and uniform noise. This study introduces a novel and effective amalgamation approach for HQCNN, laying the groundwork for the advancement and application of quantum advantages in artificial intelligence computations. [ABSTRACT FROM AUTHOR]

Robots have been widely used in daily life in recent years. Unlike conventional robots made of rigid materials, soft robots utilize stretchable and flexible materials, allowing flexible movements similar to those of living organisms, which are difficult for traditional robots. Previous studies have used periodic signals to control soft robots, which lead to repetitive motions and make it challenging to generate environment-adapted motions. To address this issue, control methods can be learned through deep reinforcement learning to enable soft robots to select appropriate actions based on observations, improving their adaptability to environmental changes. In addition, as mobile robots have limited onboard resources, it is necessary to conserve battery consumption and achieve low-power control. Therefore, the use of spiking neural networks (SNNs) with neuromorphic chips enables low-power control of soft robots. In this study, we investigated the learning methods for SNNs aimed at controlling soft robots. Experiments were conducted using a caterpillar-like soft robot model based on previous studies, and the effectiveness of the learning method was evaluated. [ABSTRACT FROM AUTHOR]

The electrocardiogram (ECG) is one of the most significant methods of diagnostics for determining heart rhythm disorders. For this study, raw ECG signals from the Physio Bank database are subjected to an important preprocessing step that uses empirical mode decomposition (EMD) on signal denoising and distortion elimination. Establishing functioning spiking neural networks (SNN) involves figuring out the neuron's state through its activity level, challenging due to its resemblance to the human brain's data processing, yet appealing due to factors like improved unsupervised learning methods, with ten parameters chosen for the learning algorithm of SNN. A comprehensive set of 15 different time-domain features and 10 Cepstral domain features is precisely extracted to train the SNN classifier. An extensive study is conducted to analyse the learning parameters that affect SNN performance, significantly influencing result accuracy. Through a two-classification process, the differentiation between normal and abnormal ECG patterns can be achieved in this study. A maximum testing accuracy of 91.6667% and a maximum training accuracy of 99.1667% have been attained by the process. These results demonstrate the competency of the system in distinguishing between distinct ECG classes, particularly in identifying normal and abnormal cardiac rhythms through ECG classification. [ABSTRACT FROM AUTHOR]

Many studies suggest that probabilistic spiking in biological neural systems is beneficial as it aids learning and provides Bayesian inference-like dynamics. If appropriately utilised, noise and stochasticity in nanoscale devices can benefit neuromorphic systems. In this paper, we build a stochastic leaky integrate and fire (LIF) neuron, utilising a Mott memristor's inherent stochastic switching dynamics. We demonstrate that the developed LIF neuron is capable of biological neural dynamics. We leverage these characteristics of the proposed LIF neuron by integrating it into a population-coded spiking neural network and a spiking restricted Boltzmann machine (sRBM), thereby showcasing its ability to implement probabilistic learning and inference. The sRBM achieves a software-comparable accuracy of 87.13%. Unlike CMOS-based probabilistic neurons, our design does not require any external noise sources. The designed neurons are highly energy efficient and ultra-compact, requiring only three components: a resistor, a capacitor and a memristor device. [ABSTRACT FROM AUTHOR]

Efficient and effective radar emitter recognition is critical for electronic support measurement (ESM) systems. However, in complex electromagnetic environments, intercepted pulse trains generally contain substantial data noise, including spurious and missing pulses. Currently, radar emitter recognition methods utilizing traditional artificial neural networks (ANNs) like CNNs and RNNs are susceptible to data noise and require intensive computations, posing challenges to meeting the performance demands of modern ESM systems. Spiking neural networks (SNNs) exhibit stronger representational capabilities compared to traditional ANNs due to the temporal dynamics of spiking neurons and richer information encoded in precise spike timing. Furthermore, SNNs achieve higher computational efficiency by performing event-driven sparse addition calculations. In this paper, a lightweight spiking neural network is proposed by combining direct coding, leaky integrate-and-fire (LIF) neurons, and surrogate gradients to recognize radar emitters. Additionally, an improved SNN for radar emitter recognition is proposed, leveraging the local timing structure of pulses to enhance adaptability to data noise. Simulation results demonstrate the superior performance of the proposed method over existing methods. [ABSTRACT FROM AUTHOR]

In this article, we propose a circuit to imitate the behavior of a Reward-Modulated spike-timing-dependent plasticity synapse. When two neurons in adjacent layers produce spikes, each spike modifies the thickness in the shared synapse. As a result, the synapse's ability to conduct impulses is controlled, leading to an unsupervised learning rule. By introducing a reward signal, reinforcement learning is enabled by redirecting the growth and shrinkage of synapses based on signal feedback from the environment. The proposed synapse manages the convolution of the emitted spike signals to promote either the strengthening or weakening of the synapse, represented as the resistance value of a memristor device. As memristors have a conductance range that may differ from the available current input range of typical CMOS neuron designs, the synapse circuit can be adjusted to regulate the spike's amplitude current to comply with the neuron. The circuit described in this work allows for the implementation of fully interconnected layers of neuron analog circuits. This is achieved by having each synapse reconform the spike signal, thus removing the burden of providing enough power from the neurons to each memristor. The synapse circuit was tested using a CMOS analog neuron described in the literature. Additionally, the article provides insight into how to properly describe the hysteresis behavior of the memristor in Verilog-A code. The testing and learning capabilities of the synapse circuit are demonstrated in simulation using the Skywater-130 nm process. The article's main goal is to provide the basic building blocks for deep neural networks relying on spiking neurons and memristors as the basic processing elements to handle spike generation, propagation, and synaptic plasticity. [ABSTRACT FROM AUTHOR]

Introduction: Great knowledge was gained about the computational substrate of the brain, but the way in which components and entities interact to perform information processing still remains a secret. Complex and large-scale network models have been developed to unveil processes at the ensemble level taking place over a large range of timescales. They challenge any kind of simulation platform, so that efficient implementations need to be developed that gain from focusing on a set of relevant models. With increasing network sizes imposed by these models, low latency inter-node communication becomes a critical aspect. This situation is even accentuated, if slow processes like learning should be covered, that require faster than real-time simulation. Methods: Therefore, this article presents two simulation frameworks, in which network-on-chip simulators are interfaced with the neuroscientific development environment NEST. This combination yields network traffic that is directly defined by the relevant neural network models and used to steer the network-on- chip simulations. As one of the outcomes, instructive statistics on network latencies are obtained. Since time stamps of different granularity are used by the simulators, a conversion is required that can be exploited to emulate an intended acceleration factor. Results: By application of the frameworks to scaled versions of the cortical microcircuit model--selected because of its unique properties as well as challenging demands--performance curves, latency, and traffic distributions could be determined. Discussion: The distinct characteristic of the second framework is its tree- based source-address driven multicast support, which, in connection with the torus topology, always led to the best results. Although currently biased by some inherent assumptions of the network-on-chip simulators, the results suit well to those of previous work dealing with node internals and suggesting accelerated simulations to be in reach. [ABSTRACT FROM AUTHOR]

Spiking Neural Networks have gained significant attention due to their potential for energy efficiency and biological plausibility. However, the reduced number of user-friendly tools for designing, training, and visualizing Spiking Neural Networks hinders widespread adoption. This paper presents the SNNtrainer3D v1.0.0, a novel software application that addresses these challenges. The application provides an intuitive interface for designing Spiking Neural Networks architectures, with features such as dynamic architecture editing, allowing users to add, remove, and edit hidden layers in real-time. A key innovation is the integration of Three.js for three-dimensional visualization of the network structure, enabling users to inspect connections and weights and facilitating a deeper understanding of the model's behavior. The application supports training on the Modified National Institute of Standards and Technology dataset and allows the downloading of trained weights for further use. Moreover, it lays the groundwork for future integration with physical memristor technology, positioning it as a crucial tool for advancing neuromorphic computing research. The advantages of the development process, technology stack, and visualization are discussed. The SNNtrainer3D represents a significant step in making Spiking Neural Networks more accessible, understandable, and easier for Artificial Intelligence researchers and practitioners. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Remote Sensing is the property of Editorial Office of Journal of Remote Sensing & Science Publishing Co. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Ants are capable of learning long visually guided foraging routes with limited neural resources. The visual scene memory needed for this behaviour is mediated by the mushroom bodies; an insect brain region important for learning and memory. In a visual navigation context, the mushroom bodies are theorised to act as familiarity detectors, guiding ants to views that are similar to those previously learned when first travelling along a foraging route. Evidence from behavioural experiments, computational studies and brain lesions all support this idea. Here we further investigate the role of mushroom bodies in visual navigation with a spiking neural network model learning complex natural scenes. By implementing these networks in GeNN-a library for building GPU accelerated spiking neural networks-we were able to test these models offline on an image database representing navigation through a complex outdoor natural environment, and also online embodied on a robot. The mushroom body model successfully learnt a large series of visual scenes (400 scenes corresponding to a 27 m route) and used these memories to choose accurate heading directions during route recapitulation in both complex environments. Through analysing our model's Kenyon cell (KC) activity, we were able to demonstrate that KC activity is directly related to the respective novelty of input images. Through conducting a parameter search we found that there is a non-linear dependence between optimal KC to visual projection neuron (VPN) connection sparsity and the length of time the model is presented with an image stimulus. The parameter search also showed training the model on lower proportions of a route generally produced better accuracy when testing on the entire route. We embodied the mushroom body model and comparator visual navigation algorithms on a Quanser Q-car robot with all processing running on an Nvidia Jetson TX2. On a 6.5 m route, the mushroom body model had a mean distance to training route (error) of 0.144 ±0.088 m over 5 trials, which was performance comparable to standard visual-only navigation algorithms. Thus, we have demonstrated that a biologically plausible model of the ant mushroom body can navigate complex environments both in simulation and the real world. Understanding the neural basis of this behaviour will provide insight into how neural circuits are tuned to rapidly learn behaviourally relevant information from complex environments and provide inspiration for creating bio-mimetic computer/robotic systems that can learn rapidly with low energy requirements. [ABSTRACT FROM AUTHOR]

A variant of membrane computing models called Spiking Neural P systems (SNP systems) closely mimics the structure and behavior of biological neurons. As third-generation neural networks, SNP systems have flexible architectures allowing the design of bio-inspired machine learning algorithms. This paper proposes Modular Spiking Neural P (MSNP) systems to solve image classification problems, a novel SNP system to be applied in scenarios where hundreds or even thousands of different classes are considered. A main issue to face in such situations is related to the structural complexity of the network. MSNP systems devised in this work allow to approach the general classification problem by dividing it in smaller parts, that are then faced by single entities of the network. As a benchmark dataset, the Oxford Flowers 102 dataset is considered, consisting of more than 8000 pictures of flowers belonging to the 102 species commonly found in the UK. These classes sometimes present large variations within them, may be also very similar to one another, and different images of the same subject may differ a lot. The work describes the architecture of the MSNP system, based on modules focusing on a specific class, their training phase, and the evaluation of the model both concerning result accuracy as well as energy consumption. Experimental results on image classification problems show that the model achieves good results, but is strongly connected to image quality, mainly depending on the frequency of images, remarkable changes of pose, images not centered, and subject mostly not shown. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) have superior energy efficiency due to their spiking signal transmission, which mimics biological nervous systems, but they are difficult to train effectively. Although surrogate gradient-based methods offer a workable solution, trained SNNs frequently fall into local minima because they are still primarily based on gradient dynamics. Inspired by the chaotic dynamics in animal brain learning, we propose a chaotic spiking backpropagation (CSBP) method that introduces a loss function to generate brain-like chaotic dynamics and further takes advantage of the ergodic and pseudo-random nature to make SNN learning effective and robust. From a computational viewpoint, we found that CSBP significantly outperforms current state-of-the-art methods on both neuromorphic data sets (e.g. DVS-CIFAR10 and DVS-Gesture) and large-scale static data sets (e.g. CIFAR100 and ImageNet) in terms of accuracy and robustness. From a theoretical viewpoint, we show that the learning process of CSBP is initially chaotic, then subject to various bifurcations and eventually converges to gradient dynamics, consistently with the observation of animal brain activity. Our work provides a superior core tool for direct SNN training and offers new insights into understanding the learning process of a biological brain. [ABSTRACT FROM AUTHOR]

Neuromorphic devices capable of training spiking neural networks (SNNs) are not easy to develop due to two main factors: lack of efficient supervised learning algorithms, and high computational requirements that ultimately lead to higher power consumption and higher cost. In this article, we present an FPGA-based neuromorphic system capable of training SNNs efficiently. The Tempotron learning rule along with population coding is adopted for SNN learning to achieve a high level of classification accuracy. To blend cost efficiency with high throughput, integration of both integrate-and-fire (IF) and leaky integrate-and-fire (LIF) neurons is proposed. Moreover, the post-synaptic potential (PSP) kernel function for the LIF neuron is modeled using slopes. This novel solution obviates the need for multipliers and memory accesses for kernel computations. Experimental results show that a speedup of about 15 × can be obtained on a general-purpose Von-Neumann device if the proposed scheme is adopted. Moreover, the proposed neuromorphic design is fully parallelized and can achieve a maximum throughput of about 2460 × 10 6 4-input samples per second, while consuming only 13.6 slice registers per synapse and 89.5 look-up tables (LuTs) per synapse on Virtex 6 FPGA. The system can classify an input sample in about 4.88 ns. [ABSTRACT FROM AUTHOR]

Machine unlearning, which is crucial for data privacy and regulatory compliance, involves the selective removal of specific information from a machine learning model. This study focuses on implementing machine unlearning in Spiking Neuron Models (SNMs) that closely mimic biological neural network behaviors, aiming to enhance both flexibility and ethical compliance of AI models. We introduce a novel hybrid approach for machine unlearning in SNMs, which combines selective synaptic retraining, synaptic pruning, and adaptive neuron thresholding. This methodology is designed to effectively eliminate targeted information while preserving the overall integrity and performance of the neural network. Extensive experiments were conducted on various computer vision datasets to assess the impact of machine unlearning on critical performance metrics such as accuracy, precision, recall, and ROC AUC. Our findings indicate that the hybrid approach not only maintains but in some cases enhances the neural network’s performance post-unlearning. The results confirm the practicality and efficiency of our approach, underscoring its applicability in real-world AI systems. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) can do inference with low power consumption due to their spike sparsity. Although SNNs can be combined with neuromorphic hardware to achieve efficient inference, they are often difficult to train directly due to discrete non-differentiable spikes. As an alternative, ANN-SNN conversion is an efficient way to achieve deep SNNs by converting well-trained artificial neural networks (ANNs). However, the existing methods commonly use constant threshold for conversion. A high constant threshold value prevents neurons from rapidly delivering spikes to deeper layers and causes high time delay. In addition, the same response for different inputs may result in information loss during the information transmission. Inspired by the biological adaptive threshold mechanism, we propose a multistage adaptive threshold (MSAT) method to alleviate this problem. Instead of using a single, constant value, the threshold is adjusted in multistages, adapting to each neuron's firing history and input properties. Specifically, for each neuron, the dynamic threshold is positively correlated with the average membrane potential and negatively correlated with the rate of depolarization. The adaptation to membrane potential and input allows a timely adjustment of the threshold to fire spikes faster and transmit more information. Moreover, we analyze the spikes of inactivated neurons error, which is pervasive in early time steps. We also propose spike confidence accordingly to measure confidence about the neurons that correctly deliver spikes. Such spike confidence in early time steps is used to determine whether to elicit the spike to alleviate the spikes of inactivated neurons error. Combined with the proposed methods, we examine the performance on CIFAR-10, CIFAR-100, and ImageNet datasets. We also conduct sentiment classification and speech recognition experiments on the IDBM and Google speech commands datasets, respectively. Experiments show that our methods can achieve near-lossless and lower latency ANN-SNN conversion. In summary, we build a biologically inspired multistage adaptive threshold for converted SNN, with comparable performance to state-of-the-art methods while improving energy efficiency. [ABSTRACT FROM AUTHOR]

Dynamic memristors with short-term memory have been widely researched for their high biomimetic properties to mimic the leaky and integration functions of biological neurons. However, the limited reliability and poor switching window still hinder the implementation of dynamic memristors in a neuromorphic system. Here, we propose a highly reliable dynamic memristor with nanolayered NbO2 material based on the transition effect between two-electron emission processes. The introduced additional oxygen defects effectively enhance the electron emission effect and therefore enlarge the switching window over 107, accompanied by a low variation below 2%. Meanwhile, the designed conduction transition avoids the structural damage and the direct connection between two electrodes, benefiting the high endurance (>109) and low operating energy (19 fJ). Subsequently, we confirm the conduction transition mechanism during the dynamic behavior by data analysis, device surface expansion detection, and oxygen-controlled fabrication. The proposed mechanism is mathematically verified by a constructed compact model, which can be used not only to instruct the device fabrication but also to simulate neuron design. Based on the developed neuron with dynamic memristors, we construct a spiking neural network. The interaction between different aspects of the memristor in the network is discussed, which helps to achieve a reasonable compromise among the network accuracy (94.63%), energy consumption, and peripheral circuit requirements. [ABSTRACT FROM AUTHOR]

One of today's main concerns is to bring artificial intelligence capabilities to embedded systems for edge applications. The hardware resources and power consumption required by state-of-the-art models are incompatible with the constrained environments observed in edge systems, such as IoT nodes and wearable devices. Spiking Neural Networks (SNNs) can represent a solution in this sense: inspired by neuroscience, they reach unparalleled power and resource efficiency when run on dedicated hardware accelerators. However, when designing such accelerators, the amount of choices that can be taken is huge. This paper presents SpikExplorer, a modular and flexible Python tool for hardware-oriented Automatic Design Space Exploration to automate the configuration of FPGA accelerators for SNNs. SpikExplorer enables hardware-centric multiobjective optimization, supporting target factors such as accuracy, area, latency, power, and various combinations during the exploration process. The tool searches the optimal network architecture, neuron model, and internal and training parameters leveraging Bayesian optimization, trying to reach the desired constraints imposed by the user. It allows for a straightforward network configuration, providing the full set of explored points for the user to pick the trade-off that best fits their needs. The potential of SpikExplorer is showcased using three benchmark datasets. It reaches 95.8% accuracy on the MNIST dataset, with a power consumption of 180 mW/image and a latency of 0.12 ms/image, making it a powerful tool for automatically optimizing SNNs. [ABSTRACT FROM AUTHOR]

For some years now, the locomotion mechanisms used by vertebrate animals have been a major inspiration for the improvement of robotic systems. These mechanisms range from adapting their movements to move through the environment to the ability to chase prey, all thanks to senses such as sight, hearing, and touch. Neuromorphic engineering is inspired by brain problem‐solving techniques with the goal of implementing models that take advantage of the characteristics of biological neural systems. While this is a well‐defined and explored area in this field, there is no previous work that fuses analog and neuromorphic sensors to control and modify robotic behavior in real time. Herein, a system is presented based on spiking neural networks implemented on the SpiNNaker hardware platform that receives information from both analog (force‐sensing resistor) and digital (neuromorphic retina) sensors and is able to adapt the speed and orientation of a hexapod robot depending on the stability of the terrain where it is located and the position of the target. These sensors are used to modify the behavior of different spiking central pattern generators, which in turn will adapt the speed and orientation of the robotic platform, all in real time. In particular, experiments show that the network is capable of correctly adapting to the stimuli received from the sensors, modifying the speed and heading of the robotic platform. [ABSTRACT FROM AUTHOR]