Search Results (26)

User Behaviour Analytics (UBA) relies heavily on sequential data to detect anomalies such as insider threats. Traditional approaches often model user behaviour as discrete sequences of events using Recurrent Neural Networks (RNNs) like Long Short-Term Memory (LSTM) networks. These methods implicitly treat time steps as uniform, ignoring the irregular time intervals inherent in user logs. In this paper, we present the first application of Neural Controlled Differential Equations (NCDEs) to user behaviour analytics, a class of continuous-time models that naturally handle irregularly-timed event data. We compare a simple LSTM predictor against an NCDE predictor on the CERT 4.2 and 6.2 insider threat dataset. We demonstrate that standard discrete-time models (LSTMs) produce noisy loss signals on sparse data, forcing downstream classifiers to rely on fragile error spikes. In contrast, Neural CDEs generate stable, continuous error signals. NCDE roughly tripled the F1 of the discrete baseline (0.364 vs. 0.133) on the challenging CERT 6.2 dataset. Full article

Price spikes are rare but economically significant events observed across electricity, financial, commodity, and cryptocurrency markets. Their abrupt magnitude, heavy-tailed distributions, and severe class imbalance make them difficult to forecast using conventional time-series methods. This systematic literature review, conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, synthesises recent deep learning approaches to forward-looking price-spike prediction and classification. Searches of Scopus, Web of Science, and IEEE Xplore identified studies published between 2020 and 2026. Following screening and full-text eligibility assessment of approximately 300 studies, only 20 met the inclusion criteria and were included in the final synthesis, comprising 19 peer-reviewed papers and one doctoral thesis. The review develops a structured taxonomy spanning spike definitions, task formulations, model architectures, input design, and evaluation practices. A central finding is that predictive performance is driven more by problem formulation, label construction, and evaluation design than by model architecture. While architectures have diversified to include recurrent networks, transformers, graph neural networks, and hybrid frameworks, improvements are often attributable to differences in how the prediction problem is defined rather than the models themselves. Key limitations stem from inconsistent spike definitions and insufficient treatment of class imbalance, leading to a misalignment between modelling objectives and evaluation practices, further exacerbated by the absence of standardised benchmarks. These issues hinder comparability and can lead to overstated model performance by masking poor detection of rare but economically critical spike events. The review therefore identifies clear directions for future research, including standardised spike labelling, adoption of rare-event-appropriate evaluation frameworks, and problem formulations that explicitly target extreme-event prediction. Full article

Autonomous driving perception demands low latency, high temporal resolution, and stringent hardware efficiency. While event-based spiking neural networks (SNNs) offer bio-inspired sparse computation, their deployment on edge field-programmable gate arrays (FPGAs) is obstructed by irregular execution patterns and temporal state storage overhead. To address this, we propose HAPQ, a unified hardware-aware pruning and quantization pipeline for compact event-based object detection. Starting from an end-to-end adaptive sampling SNN detector (EAS-SNN), HAPQ conducts hardware-aware configuration search within discrete digital signal processor (DSP) and block RAM (BRAM) budgets, applies single-instruction-multiple-data (SIMD)-aligned structured pruning for computational regularity, and jointly quantizes synaptic weights and membrane potentials via a shift-friendly fixed-point recurrence. Evaluation on the Prophesee Gen1 dataset and an FPGA accelerator shows that HAPQ improves detection accuracy from 0.284 to 0.425 in mean average precision ($mA{P}_{50:95}$) and achieves 0.722 $mA{P}_{50}$. Hardware implementation reveals a reduction in lookup table (LUT) usage to 1680, complete DSP elimination, and a maximum operating frequency of 920.81 MHz at 0.630 W. These results confirm that effective temporal SNN deployment requires joint optimization of model architecture, state precision, and hardware-aligned workload organization. Full article

Spiking neural networks (SNNs) support energy-efficient machine intelligence because event-driven computation and sparse activity map naturally to low-power digital hardware. In practical implementations, however, membrane states, synaptic weights, and thresholds are represented with finite-precision integer arithmetic. Quantization, clipping, and overflow can therefore alter network dynamics rather than merely approximate a higher-precision model. This paper adopts an integer-state dynamical perspective, modeling a quantized SNN with a hardware-relevant update rule as a deterministic map on a bounded integer lattice. Rather than claiming recurrence itself as a new property, we focus on how finite-precision representation and implementation semantics shape observed recurrent regimes and activity patterns. We introduce a shift-based update rule with integer-valued states and investigate its behaviour through simulation-based analysis with network sizes $N=30$–130, connection densities 0.1–0.9, and bit widths 1 to 16 over T = 1000 steps. The results show bounded and recurrent temporal structure with strong quantization sensitivity. The observed regimes depend heavily on the semantics of representation and the scaling choices. These findings suggest that numerical precision can act as a dynamical design variable and provide useful implications for hardware-oriented SNN design, while motivating future work on attractor analysis and FPGA/ASIC validation. Full article

Liquid State Machines (LSMs), a reservoir computing model based on recurrent spiking neural networks, provide a powerful framework for solving spatiotemporal classification tasks by leveraging rich temporal dynamics and event-driven processing. Although the traditional LSM formulation assumes a fixed, randomly generated reservoir, recent research has explored optimization strategies to improve liquid dynamics. However, most existing approaches focus primarily on optimizing synaptic connectivity or reservoir structure, while the role of neuron-level parameters remains largely underexplored. This work proposes a neuroevolutionary strategy based on a Genetic Algorithm (GA) that encodes both neuron configurations and their spatial positions, explicitly treating neuron-level parameters as optimization targets. By evolving neuron-specific parameters and spatial positions, the method induces diverse reservoir dynamics. Unlike approaches that directly optimize synaptic weights, the proposed representation maintains an encoding whose dimensionality scales linearly with the number of neurons. The approach was evaluated on four synthetic benchmark tasks, including one Frequency Recognition task and three Pattern Recognition tasks, using compact reservoirs composed of only 20 Leaky Integrate-and-Fire neurons. Despite the small reservoir size, the method achieved state-of-the-art or highly competitive performance, reaching mean accuracies of up to 99.71%. In the most challenging case (PR12), performance improved when the reservoir size was increased to 64 neurons. The method was further evaluated on two real-world datasets, N-MNIST and the Free Spoken Digit Dataset (FSDD), using reservoirs of 300 neurons, achieving 90.65% and 81.47% accuracy, respectively, while using substantially fewer neurons than many existing LSM-based approaches. These results highlight the potential of evolving neuron configurations and spatial organization to produce compact and effective liquid reservoirs. Full article

Precise Estimated Time of Arrival (ETA) prediction in Terminal Maneuvering Areas (TMA) constitutes a prerequisite for efficient arrival sequencing and airspace capacity management. While data-driven approaches outperform kinematic models, conventional Recurrent Neural Networks (RNNs) exhibit limitations in modeling complex multi-aircraft spatial interactions and lack the capability to quantify predictive uncertainty. Conversely, Spiking Neural Networks (SNNs) enable energy-efficient event-driven computation, yet their applicability to continuous trajectory regression is hindered by “input starvation,” where normalized state vectors fail to induce sufficient neural firing rates. This study proposes a Probabilistic Spiking Transformer (PST) architecture to integrate neuromorphic sparsity with global attention mechanisms. An Adaptive Spiking Temporal Encoding mechanism incorporating learnable linear projections is introduced to resolve the regression-spiking incompatibility, facilitating the autonomous mapping of continuous trajectory dynamics into sparse spike trains without heuristic scaling. Concurrently, a Distance-Biased Multi-Aircraft Cross-Attention (MACA) module models air traffic conflicts by weighting spatial interactions according to physical proximity, thereby embedding separation constraints into the feature extraction process. Evaluation on large-scale real-world ADS-B datasets demonstrates that the PST yields a Mean Absolute Error (MAE) of 49.27 s, representing a 60% error reduction relative to standard LSTM baselines. Furthermore, the model generates well-calibrated probabilistic distributions (Prediction Interval Coverage Probability > 94%), offering quantifiable uncertainty metrics for risk-based decision support while ensuring real-time inference suitable for operational deployment. Full article

Reservoir computing (RC) has emerged as an energy-efficient paradigm for temporal information processing, offering reduced training complexity by fixing recurrent dynamics and training only a simple readout layer. Among RC models, Echo State Networks (ESNs) and Liquid State Machines (LSMs) represent two distinct approaches based on continuous-valued and spiking neural dynamics, respectively. In this work, we present a comparative evaluation of ESNs and LSMs on the Mackey–Glass chaotic time-series prediction task, with emphasis on scalability, overfitting behavior, and robustness to reduced numerical error precision. Experimental results show that ESNs achieve lower prediction error with relatively small reservoirs but exhibit early performance saturation and signs of overfitting as reservoir size increases. In contrast, LSMs demonstrate more consistent generalization with increasing reservoir size and maintain stable performance under aggressive reservoir quantization. These findings highlight fundamental trade-offs between accuracy and hardware efficiency, and suggest that spiking RC models are well suited for energy-constrained and neuromorphic computing applications. Full article

Background/Objectives: From birth, infants learn how to interact with the world through exploration. It has been proposed that this early learning phase is driven by motor babbling: the spontaneous generation of exploratory movements that are progressively consolidated through associative mechanisms. This process leads to the acquisition of a repertoire of hand movements such as single- or multi-finger flexion, extension, touching, and pushing. Later, in a second phase, some of these movements (e.g., those that happen to enable access to biologically salient stimuli, such as grasping food) are further reinforced and consolidated through rewards obtained from the environment. However, the neural mechanisms underlying these processes remain unclear. Here, we used a fully neuroanatomically and neurophysiologically constrained neural network model to investigate the brain correlates of these processes. Methods: The model consists of six neural maps simulating six human brain areas, including three pre-central (motor-related) and three post-central (sensory-related) regions. Each map is composed of excitatory and inhibitory spiking neurons, with biologically constrained within- and between-area connectivity forming recurrent circuits. Hand action execution and corresponding haptic perception are simulated simply as activity in primary motor and somatosensory model areas, respectively. During an initial “exploratory” phase, the network learned, via Hebbian mechanisms, associations—as emerging distributed cell assembly (CA) circuits—linking “motor” to corresponding “haptic feedback” patterns. As a result of this initial training, the model began to exhibit spontaneous ignitions of these CA circuits, an emergent phenomenon taken to represent internally generated, non-stimulus-driven attempts at hand action exploitation. In a second phase, a global reward signal, simulating dopamine-mediated reward encoding, was applied to only a subset of “successful” actions upon their noise-driven ignition. Results: During the first exploratory phase, the neural architecture autonomously developed “action-perception” circuits corresponding to multiple possible hand actions. During the subsequent exploitation phase, positively reinforced circuits increased in size and, consequently, in frequency of spontaneous ignition, when compared to non-rewarded “actions”. Conclusions: These results provide a mechanistic account, at the cortical-circuit level, of the early acquisition of hand actions, of their subsequent consolidation, and of the spontaneous transition of an agent’s behavior from exploration to reward-seeking, as typically observed in humans and animals during development. Full article

The detection of forced oscillations, especially distinguishing them from natural oscillations, has emerged as a major concern in power system stability monitoring. Deep learning (DL) holds significant potential for detecting forced oscillations correctly. However, existing artificial neural networks (ANNs) face challenges when employed in edge devices for timely detection due to their inherent complex computations and high power consumption. This paper proposes a novel hybrid network that integrates a spiking recurrent neural network (SRNN) with long short-term memory (LSTM). The SRNN achieves computational and energy efficiency, while the integration with LSTM is conducive to effectively capturing temporal dependencies in time-series oscillation data. The proposed hybrid network is trained using the backpropagation-through-time (BPTT) optimization algorithm, with adjustments made to address the discontinuous gradient in the SRNN. We evaluate our proposed model on both simulated and real-world oscillation datasets. Overall, the experimental results demonstrate that the proposed model can achieve higher accuracy and superior performance in distinguishing forced oscillations from natural oscillations, even in the presence of strong noise, compared to pure LSTM and other SRNN-related models. Full article

Current technological solutions for Brain-machine Interfaces (BMI) achieve reasonable accuracy, but most systems are large in size, power consuming and not auto-adaptive. This work addresses the question whether current neuromorphic technologies could resolve these problems? The paper proposes a novel neuromorphic framework of a BMI system for prosthetics control via decoding Electro Cortico-Graphic (ECoG) brain signals. It includes a three-dimensional spike timing neural network (3D-SNN) for brain signals features extraction and an on-line trainable recurrent reservoir structure (Echo state network (ESN)) for Motor Control Decoding (MCD). A software system, written in Python using NEST Simulator SNN library is described. It is able to adapt continuously in real time in supervised or unsupervised mode. The proposed approach was tested on several experimental data sets acquired from a tetraplegic person. First simulation results are encouraging, showing also the need for a further improvement via multiple hyper-parameters tuning. Its future implementation on a neuromorphic hardware platform that is smaller in size and significantly less power consuming is discussed too. Full article

This study compares three machine learning and deep learning models—Support Vector Regression (SVR), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM)—for predicting market prices using the NGX All-Share Index dataset. The models were evaluated using multiple error metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE), Mean Percentage Error (MPE), and R-squared. RNN and LSTM were tested with both 30 and 60-day windows, with performance compared to SVR. LSTM delivered better R-squared values, with a 60-day LSTM achieving the best accuracy (R-squared = 0.993) when using a combination of endogenous market data and technical indicators. SVR showed reliable results in certain scenarios but struggled in fold 2 with a sudden spike that shows a high probability of not capturing the entire underlying NGX pattern in the dataset correctly, as witnessed by the high validation loss during the period. Additionally, RNN faced the vanishing gradient problem that limits its long-term performance. Despite challenges, LSTM’s ability to handle temporal dependencies, especially with the inclusion of On-Balance Volume, led to significant improvements in prediction accuracy. The use of the Optuna optimisation framework further enhanced model training and hyperparameter tuning, contributing to the performance of the LSTM model. Full article

Object segmentation in images is typically spatial and focuses on the spatial coherence of pixels. Nanoparticles in electron microscopy images are also segmented frame by frame, with subsequent morphological analysis. However, morphological analysis is inherently sequential, and a temporal regularity is evident in the process. In this study, we extend the spatially focused morphological analysis by incorporating a fusion of hard and soft inductive bias from sequential machine learning techniques to account for temporal relationships. Previously, spiky Au nanoparticles (Au-SNPs) in electron microscopy images were analyzed, and their morphological properties were automatically generated using a hourglass convolutional neural network architecture. In this study, recurrent layers are integrated to capture the natural, sequential growth of the particles. The network is trained with a spike-focused loss function. Continuous segmentation of the images explores the regressive relationships among natural growth features, generating morphological statistics of the nanoparticles. This study comprehensively evaluates the proposed approach by comparing the results of segmentation and morphological properties analysis, demonstrating its superiority over earlier methods. Full article

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. Full article

Addressing common challenges such as limited indicators, poor adaptability, and imprecise modeling in gas pre-warning systems for driving faces, this study proposes a hybrid predictive and pre-warning model grounded in time-series analysis. The aim is to tackle the effects of broad application across diverse mines and insufficient data on warning accuracy. Firstly, we introduce an adaptive normalization (AN) model for standardizing gas sequence data, prioritizing recent information to better capture the time-series characteristics of gas readings. Coupled with the Gated Recurrent Unit (GRU) model, AN demonstrates superior forecasting performance compared to other standardization techniques. Next, Ensemble Empirical Mode Decomposition (EEMD) is used for feature extraction, guiding the selection of the Variational Mode Decomposition (VMD) order. Minimal decomposition errors validate the efficacy of this approach. Furthermore, enhancements to the transformer framework are made to manage non-linearities, overcome gradient vanishing, and effectively analyze long time-series sequences. To boost versatility across different mining scenarios, the Optuna framework facilitates multiparameter optimization, with xgbRegressor employed for accurate error assessment. Predictive outputs are benchmarked against Recurrent Neural Networks (RNN), GRU, Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM), where the hybrid model achieves an R-squared value of 0.980975 and a Mean Absolute Error (MAE) of 0.000149, highlighting its top performance. To cope with data scarcity, bootstrapping is applied to estimate the confidence intervals of the hybrid model. Dimensional analysis aids in creating real-time, relative gas emission metrics, while persistent anomaly detection monitors sudden time-series spikes, enabling unsupervised early alerts for gas bursts. This model demonstrates strong predictive prowess and effective pre-warning capabilities, offering technological reinforcement for advancing intelligent coal mine operations. Full article

Efficient energy use is crucial for achieving carbon neutrality and reduction. As part of these efforts, research is being carried out to apply a phase change material (PCM) to a concrete structure together with an aggregate. In this study, an energy consumption simulation was performed using data from concrete mock-up structures. To perform the simulation, the threshold investigation was performed through the Bayesian approach. Furthermore, the spiking part of the spiking neural network was modularized and integrated into a recurrent neural network (RNN) to find accurate energy consumption. From the training-test results of the trained neural network, it was possible to predict data with an R² value of 0.95 or higher through data prediction with high accuracy for the RNN. In addition, the spiked parts were obtained; it was found that PCM-containing concrete could consume 32% less energy than normal concrete. This result suggests that the use of PCM can be a key to reducing the energy consumption of concrete structures. Furthermore, the approach of this study is considered to be easily applicable in energy-related institutions and the like for predicting energy consumption during the summer. Full article