Real-time hardware emulation of neural cultures: A comparative study of in vitro, in silico and in duris silico models.

Biological neural networks are well known for their capacity to process information with extremely low power consumption. Fields such as Artificial Intelligence, with high computational costs, are seeking for alternatives inspired in biological systems. An inspiring alternative is to implement hardware architectures that replicate the behavior of biological neurons but with the flexibility in programming capabilities of an electronic device, all combined with a relatively low operational cost. To advance in this quest, here we analyze the capacity of the HEENS hardware architecture to operate in a similar manner as an in vitro neuronal network grown in the laboratory. For that, we considered data of spontaneous activity in living neuronal cultures of about 400 neurons and compared their collective dynamics and functional behavior with those obtained from direct numerical simulations (in silico) and hardware implementations (in duris silico). The results show that HEENS is capable to mimic both the in vitro and in silico systems with high efficient-cost ratio, and on different network topological designs. Our work shows that compact low-cost hardware implementations are feasible, opening new avenues for future, highly efficient neuromorphic devices and advanced human-machine interfacing.

Auto-Spikformer: Spikformer architecture search.

Introduction: The integration of self-attention mechanisms into Spiking Neural Networks (SNNs) has garnered considerable interest in the realm of advanced deep learning, primarily due to their biological properties. Recent advancements in SNN architecture, such as Spikformer, have demonstrated promising outcomes. However, we observe that Spikformer may exhibit excessive energy consumption, potentially attributable to redundant channels and blocks. Methods: To mitigate this issue, we propose a one-shot Spiking Transformer Architecture Search method, namely Auto-Spikformer. Auto-Spikformer extends the search space to include both transformer architecture and SNN inner parameters. We train and search the supernet based on weight entanglement, evolutionary search, and the proposed Discrete Spiking Parameters Search (DSPS) methods. Benefiting from these methods, the performance of subnets with weights inherited from the supernet without even retraining is comparable to the original Spikformer. Moreover, we propose a new fitness function aiming to find a Pareto optimal combination balancing energy consumption and accuracy. Results and discussion: Our experimental results demonstrate the effectiveness of Auto-Spikformer, which outperforms the original Spikformer and most CNN or ViT models with even fewer parameters and lower energy consumption.

Spike-Based Neuromorphic Hardware for Dynamic Tactile Perception with a Self-Powered Mechanoreceptor Array.

A self-powered mechanoreceptor array is demonstrated using four mechanoreceptor cells for recognition of dynamic touch gestures. Each cell consists of a triboelectric nanogenerator (TENG) for touch sensing and a bi-stable resistor (biristor) for spike encoding. It produces informative spike signals by sensing a force of an external touch and encoding the force into the number of spikes. An array of the mechanoreceptor cells is utilized to monitor various touch gestures and it successfully generated spike signals corresponding to all the gestures. To validate the practicality of the mechanoreceptor array, a spiking neural network (SNN), highly attractive for power consumption compared to the conventional von Neumann architecture, is used for the identification of touch gestures. The measured spiking signals are reflected as inputs for the SNN simulations. Consequently, touch gestures are classified with a high accuracy rate of 92.5%. The proposed mechanoreceptor array emerges as a promising candidate for a building block of tactile in-sensor computing in the era of the Internet of Things (IoT), due to the low cost and high manufacturability of the TENG. This eliminates the need for a power supply, coupled with the intrinsic high throughput of the Si-based biristor employing complementary metal-oxide-semiconductor (CMOS) technology.

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.

Self-architectural knowledge distillation for spiking neural networks.

Spiking neural networks (SNNs) have attracted attention due to their biological plausibility and the potential for low-energy applications on neuromorphic hardware. Two mainstream approaches are commonly used to obtain SNNs, i.e., ANN-to-SNN conversion methods, and Directly-trained-SNN methods. However, the former achieve excellent performance at the cost of a large number of time steps (i.e., latency), while the latter exhibit lower latency but suffers from suboptimal performance. To tackle the performance-latency trade-off, we propose Self-Architectural Knowledge Distillation (SAKD), an intuitive and effective method for SNNs leveraging Knowledge Distillation (KD). We adopt a bilevel teacher-student training strategy in SAKD, i.e., level-1 involves directly transferring same-architectural pre-trained ANN weights to SNNs, and level-2 encourages the SNNs to mimic ANN's behavior, considering both final responses and intermediate features aspects. Learning with informative supervision signals fostered by labels and ANNs, our SAKD achieves new state-of-the-art (SOTA) performance with a few time steps on widely-used classification benchmark datasets. On ImageNet-1K, with only 4 time steps, our Spiking-ResNet34 model attains a Top-1 accuracy of 70.04%, outperforming the previous same-architectural SOTA methods. Notably, our SEW-ResNet152 model reaches a Top-1 accuracy of 77.30% on ImageNet-1K, setting a new SOTA benchmark for SNNs. Furthermore, we apply our SAKD to various dense prediction downstream tasks, such as object detection and semantic segmentation, demonstrating strong generalization ability and superior performance. In conclusion, our proposed SAKD framework presents a promising approach for achieving both high performance and low latency in SNNs, potentially paving the way for future advancements in the field.

Evolution, diversity, and function of the disease susceptibility gene Snn1 in wheat.

Septoria nodorum blotch (SNB), caused by Parastagonospora nodorum, is a disease of durum and common wheat initiated by the recognition of pathogen-produced necrotrophic effectors (NEs) by specific wheat genes. The wheat gene Snn1 was previously cloned, and it encodes a wall-associated kinase that directly interacts with the NE SnTox1 leading to programmed cell death and ultimately the development of SNB. Here, sequence analysis of Snn1 from 114 accessions including diploid, tetraploid, and hexaploid wheat species revealed that some wheat lines possess two copies of Snn1 (designated Snn1-B1 and Snn1-B2) approximately 120 kb apart. Snn1-B2 evolved relatively recently as a paralog of Snn1-B1, and both genes have undergone diversifying selection. Three point mutations associated with the formation of the first SnTox1-sensitive Snn1-B1 allele from a primitive wild wheat were identified. Four subsequent and independent SNPs, three in Snn1-B1 and one in Snn1-B2, converted the sensitive alleles to insensitive forms. Protein modeling indicated these four mutations could abolish Snn1-SnTox1 compatibility either through destabilization of the Snn1 protein or direct disruption of the protein-protein interaction. A high-throughput marker was developed for the absent allele of Snn1, and it was 100% accurate at predicting SnTox1-insensitive lines in both durum and spring wheat. Results of this study increase our understanding of the evolution, diversity, and function of Snn1-B1 and Snn1-B2 genes and will be useful for marker-assisted elimination of these genes for better host resistance.

Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine.

The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests new methods for learning and identifying diagnostic biomarkers using three prominent deep learning neural network models: deep BiLSTM, reservoir SNN, and NeuCube. EEG data from datasets related to epilepsy, migraine, and healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while reservoir SNN activities and NeuCube spiking dynamics identify EEG channels as diagnostic biomarkers. BiLSTM and reservoir SNN achieve 90 and 85% classification accuracy, while NeuCube achieves 97%, all methods pinpointing potential biomarkers like T6, F7, C4, and F8. The research bears implications for refining online EEG classification, analysis, and early brain state diagnosis, enhancing AI models with interpretability and discovery. The proposed techniques hold promise for streamlined brain-computer interfaces and clinical applications, representing a significant advancement in pattern discovery across the three most popular neural network methods for addressing a crucial problem. Further research is planned to study how early can these diagnostic biomarkers predict an onset of brain states.

Neuromorphic computing spiking neural network edge detection model for content based image retrieval.

In contemporary times, content-based image retrieval (CBIR) techniques have gained widespread acceptance as a means for end-users to discern and extract specific image content from vast repositories. However, it is noteworthy that a substantial majority of CBIR studies continue to rely on linear methodologies such as gradient-based and derivative-based edge detection techniques. This research explores the integration of bioinspired Spiking Neural Network (SNN) based edge detection within CBIR. We introduce an innovative, computationally efficient SNN-based approach designed explicitly for CBIR applications, outperforming existing SNN models by reducing computational overhead by 2.5 times. The proposed SNN-based edge detection approach is seamlessly incorporated into three distinct CBIR techniques, each employing conventional edge detection methodologies including Sobel, Canny, and image derivatives. Rigorous experimentation and evaluations are carried out utilizing the Corel-10k dataset and crop weed dataset, a widely recognized and frequently adopted benchmark dataset in the realm of image analysis. Importantly, our findings underscore the enhanced performance of CBIR methodologies integrating the proposed SNN-based edge detection approach, with an average increase in mean precision values exceeding 3%. This study conclusively demonstrated the utility of our proposed methodology in optimizing feature extraction, thereby establishing its pivotal role in advancing edge centric CBIR approaches.

Bi-sigmoid spike-timing dependent plasticity learning rule for magnetic tunnel junction-based SNN.

In this study, we explore spintronic synapses composed of several Magnetic Tunnel Junctions (MTJs), leveraging their attractive characteristics such as endurance, nonvolatility, stochasticity, and energy efficiency for hardware implementation of unsupervised neuromorphic systems. Spiking Neural Networks (SNNs) running on dedicated hardware are suitable for edge computing and IoT devices where continuous online learning and energy efficiency are important characteristics. We focus in this work on synaptic plasticity by conducting comprehensive electrical simulations to optimize the MTJ-based synapse design and find the accurate neuronal pulses that are responsible for the Spike Timing Dependent Plasticity (STDP) behavior. Most proposals in the literature are based on hardware-independent algorithms that require the network to store the spiking history to be able to update the weights accordingly. In this work, we developed a new learning rule, the Bi-Sigmoid STDP (B2STDP), which originates from the physical properties of MTJs. This rule enables immediate synaptic plasticity based on neuronal activity, leveraging in-memory computing. Finally, the integration of this learning approach within an SNN framework leads to a 91.71% accuracy in unsupervised image classification, demonstrating the potential of MTJ-based synapses for effective online learning in hardware-implemented SNNs.

High-performance deep spiking neural networks via at-most-two-spike exponential coding.

Spiking neural networks (SNNs) provide necessary models and algorithms for neuromorphic computing. A popular way of building high-performance deep SNNs is to convert ANNs to SNNs, taking advantage of advanced and well-trained ANNs. Here we propose an ANN to SNN conversion methodology that uses a time-based coding scheme, named At-most-two-spike Exponential Coding (AEC), and a corresponding AEC spiking neuron model for ANN-SNN conversion. AEC neurons employ quantization-compensating spikes to improve coding accuracy and capacity, with each neuron generating up to two spikes within the time window. Two exponential decay functions with tunable parameters are proposed to represent the dynamic encoding thresholds, based on which pixel intensities are encoded into spike times and spike times are decoded into pixel intensities. The hyper-parameters of AEC neurons are fine-tuned based on the loss function of SNN-decoded values and ANN-activation values. In addition, we design two regularization terms for the number of spikes, providing the possibility to achieve the best trade-off between accuracy, latency and power consumption. The experimental results show that, compared to other similar methods, the proposed scheme not only obtains deep SNNs with higher accuracy, but also has more significant advantages in terms of energy efficiency and inference latency. More details can be found at https://github.com/RPDS2020/AEC.git.

FPGA-based fast bin-ratio spiking ensemble network for radioisotope identification.

In this work, we demonstrate the training, conversion, and implementation flow of an FPGA-based bin-ratio ensemble spiking neural network applied for radioisotope identification. The combination of techniques including learned step quantisation (LSQ) and pruning facilitated the implementation by compressing the network's parameters down to 30% yet retaining the accuracy of 97.04% with an accuracy loss of less than 1%. Meanwhile, the proposed ensemble network of 20 3-layer spiking neural networks (SNNs), which incorporates 1160 spiking neurons, only needs 334 µs for a single inference with the given clock frequency of 100 MHz. Under such optimisation, this FPGA implementation in an Artix-7 board consumes 157 µJ per inference by estimation.

A universal ANN-to-SNN framework for achieving high accuracy and low latency deep Spiking Neural Networks.

Spiking Neural Networks (SNNs) have become one of the most prominent next-generation computational models owing to their biological plausibility, low power consumption, and the potential for neuromorphic hardware implementation. Among the various methods for obtaining available SNNs, converting Artificial Neural Networks (ANNs) into SNNs is the most cost-effective approach. The early challenges in ANN-to-SNN conversion work revolved around the susceptibility of converted SNNs to conversion errors. Some recent endeavors have attempted to mitigate these conversion errors by altering the original ANNs. Despite their ability to enhance the accuracy of SNNs, these methods lack generality and cannot be directly applied to convert the majority of existing ANNs. In this paper, we present a framework named DNISNM for converting ANN to SNN, with the aim of addressing conversion errors arising from differences in the discreteness and asynchrony of network transmission between ANN and SNN. The DNISNM consists of two mechanisms, Data-based Neuronal Initialization (DNI) and Signed Neuron with Memory (SNM), designed to respectively address errors stemming from discreteness and asynchrony disparities. This framework requires no additional modifications to the original ANN and can result in SNNs with improved accuracy performance, simultaneously ensuring universality, high precision, and low inference latency. We verify it experimentally on challenging object recognition datasets, including CIFAR10, CIFAR100, and ImageNet-1k. Experimental results show that the SNN converted by our framework has very high accuracy even at extremely low latency.

Accelerating spiking neural network simulations with PymoNNto and PymoNNtorch.

Spiking neural network simulations are a central tool in Computational Neuroscience, Artificial Intelligence, and Neuromorphic Engineering research. A broad range of simulators and software frameworks for such simulations exist with different target application areas. Among these, PymoNNto is a recent Python-based toolbox for spiking neural network simulations that emphasizes the embedding of custom code in a modular and flexible way. While PymoNNto already supports GPU implementations, its backend relies on NumPy operations. Here we introduce PymoNNtorch, which is natively implemented with PyTorch while retaining PymoNNto's modular design. Furthermore, we demonstrate how changes to the implementations of common network operations in combination with PymoNNtorch's native GPU support can offer speed-up over conventional simulators like NEST, ANNarchy, and Brian 2 in certain situations. Overall, we show how PymoNNto's modular and flexible design in combination with PymoNNtorch's GPU acceleration and optimized indexing operations facilitate research and development of spiking neural networks in the Python programming language.

Benchmarking Artificial Neural Network Architectures for High-Performance Spiking Neural Networks.

Organizations managing high-performance computing systems face a multitude of challenges, including overarching concerns such as overall energy consumption, microprocessor clock frequency limitations, and the escalating costs associated with chip production. Evidently, processor speeds have plateaued over the last decade, persisting within the range of 2 GHz to 5 GHz. Scholars assert that brain-inspired computing holds substantial promise for mitigating these challenges. The spiking neural network (SNN) particularly stands out for its commendable power efficiency when juxtaposed with conventional design paradigms. Nevertheless, our scrutiny has brought to light several pivotal challenges impeding the seamless implementation of large-scale neural networks (NNs) on silicon. These challenges encompass the absence of automated tools, the need for multifaceted domain expertise, and the inadequacy of existing algorithms to efficiently partition and place extensive SNN computations onto hardware infrastructure. In this paper, we posit the development of an automated tool flow capable of transmuting any NN into an SNN. This undertaking involves the creation of a novel graph-partitioning algorithm designed to strategically place SNNs on a network-on-chip (NoC), thereby paving the way for future energy-efficient and high-performance computing paradigms. The presented methodology showcases its effectiveness by successfully transforming ANN architectures into SNNs with a marginal average error penalty of merely 2.65%. The proposed graph-partitioning algorithm enables a 14.22% decrease in inter-synaptic communication and an 87.58% reduction in intra-synaptic communication, on average, underscoring the effectiveness of the proposed algorithm in optimizing NN communication pathways. Compared to a baseline graph-partitioning algorithm, the proposed approach exhibits an average decrease of 79.74% in latency and a 14.67% reduction in energy consumption. Using existing NoC tools, the energy-latency product of SNN architectures is, on average, 82.71% lower than that of the baseline architectures.

Digital design of a spatial-pow-STDP learning block with high accuracy utilizing pow CORDIC for large-scale image classifier spatiotemporal SNN.

The paramount concern of highly accurate energy-efficient computing in machines with significant cognitive capabilities aims to enhance the accuracy and efficiency of bio-inspired Spiking Neural Networks (SNNs). This paper addresses this main objective by introducing a novel spatial power spike-timing-dependent plasticity (Spatial-Pow-STDP) learning rule as a digital block with high accuracy in a bio-inspired SNN model. Motivated by the demand for precise and accelerated computation that reduces high-cost resources in neural network applications, this paper presents a methodology based on COordinate Rotation DIgital Computer (CORDIC) definitions. The proposed designs of CORDIC algorithms for exponential (Exp CORDIC), natural logarithm (Ln CORDIC), and arbitrary power function (Pow CORDIC) are meticulously detailed and evaluated to ensure optimal acceleration and accuracy, which respectively show average errors near 10-9, 10-6, and 10-5 with 4, 4, and 6 iterations. The engineered architectures for the Exp, Ln, and Pow CORDIC implementations are illustrated and assessed, showcasing the efficiency achieved through high frequency, leading to the introduction of a Spatial-Pow-STDP learning block design based on Pow CORDIC that facilitates efficient and accurate hardware computation with 6.93 × 10-3 average error with 9 iterations. The proposed learning mechanism integrates this structure into a large-scale spatiotemporal SNN consisting of three layers with reduced hyper-parameters, enabling unsupervised training in an event-based paradigm using excitatory and inhibitory synapses. As a result, the application of the developed methodology and equations in the computational SNN model for image classification reveals superior accuracy and convergence speed compared to existing spiking networks by achieving up to 97.5%, 97.6%, 93.4%, and 93% accuracy, respectively, when trained on the MNIST, EMNIST digits, EMNIST letters, and CIFAR10 datasets with 6, 2, 2, and 6 training epochs.

DF-dRVFL: A novel deep feature based classifier for breast mass classification.

Amongst all types of cancer, breast cancer has become one of the most common cancers in the UK threatening millions of people's health. Early detection of breast cancer plays a key role in timely treatment for morbidity reduction. Compared to biopsy, which takes tissues from the lesion for further analysis, image-based methods are less time-consuming and pain-free though they are hampered by lower accuracy due to high false positivity rates. Nevertheless, mammography has become a standard screening method due to its high efficiency and low cost with promising performance. Breast mass, as the most palpable symptom of breast cancer, has received wide attention from the community. As a result, the past decades have witnessed the speeding development of computer-aided systems that are aimed at providing radiologists with useful tools for breast mass analysis based on mammograms. However, the main issues of these systems include low accuracy and require enough computational power on a large scale of datasets. To solve these issues, we developed a novel breast mass classification system called DF-dRVFL. On the public dataset DDSM with more than 3500 images, our best model based on deep random vector functional link network showed promising results through five-cross validation with an averaged AUC of 0.93 and an average accuracy of 81.71%. Compared to sole deep learning based methods, average accuracy has increased by 0.38. Compared with the state-of-the-art methods, our method showed better performance considering the number of images for evaluation and the overall accuracy.

Spiking neural networks fine-tuning for brain image segmentation.

Introduction: The field of machine learning has undergone a significant transformation with the progress of deep artificial neural networks (ANNs) and the growing accessibility of annotated data. ANNs usually require substantial power and memory usage to achieve optimal performance. Spiking neural networks (SNNs) have recently emerged as a low-power alternative to ANNs due to their sparsity nature. Despite their energy efficiency, SNNs are generally more difficult to be trained than ANNs. Methods: In this study, we propose a novel three-stage SNN training scheme designed specifically for segmenting human hippocampi from magnetic resonance images. Our training pipeline starts with optimizing an ANN to its maximum capacity, then employs a quick ANN-SNN conversion to initialize the corresponding spiking network. This is followed by spike-based backpropagation to fine-tune the converted SNN. In order to understand the reason behind performance decline in the converted SNNs, we conduct a set of experiments to investigate the output scaling issue. Furthermore, we explore the impact of binary and ternary representations in SNN networks and conduct an empirical evaluation of their performance through image classification and segmentation tasks. Results and discussion: By employing our hybrid training scheme, we observe significant advantages over both ANN-SNN conversion and direct SNN training solutions in terms of segmentation accuracy and training efficiency. Experimental results demonstrate the effectiveness of our model in achieving our design goals.

Spike Optimization to Improve Properties of Ferroelectric Tunnel Junction Synaptic Devices for Neuromorphic Computing System Applications.

The continuous advancement of Artificial Intelligence (AI) technology depends on the efficient processing of unstructured data, encompassing text, speech, and video. Traditional serial computing systems based on the von Neumann architecture, employed in information and communication technology development for decades, are not suitable for the concurrent processing of massive unstructured data tasks with relatively low-level operations. As a result, there arises a pressing need to develop novel parallel computing systems. Recently, there has been a burgeoning interest among developers in emulating the intricate operations of the human brain, which efficiently processes vast datasets with remarkable energy efficiency. This has led to the proposal of neuromorphic computing systems. Of these, Spiking Neural Networks (SNNs), designed to closely resemble the information processing mechanisms of biological neural networks, are subjects of intense research activity. Nevertheless, a comprehensive investigation into the relationship between spike shapes and Spike-Timing-Dependent Plasticity (STDP) to ensure efficient synaptic behavior remains insufficiently explored. In this study, we systematically explore various input spike types to optimize the resistive memory characteristics of Hafnium-based Ferroelectric Tunnel Junction (FTJ) devices. Among the various spike shapes investigated, the square-triangle (RT) spike exhibited good linearity and symmetry, and a wide range of weight values could be realized depending on the offset of the RT spike. These results indicate that the spike shape serves as a crucial indicator in the alteration of synaptic connections, representing the strength of the signals.