Search Results (2,193)

Neural and synaptic models that are relatively biologically accurate are easy to use to run efficient and distributed programs. The mechanism described in this paper uses these models to develop cell assemblies with a small number of neurons that persist indefinitely unless stopped. These in turn can be used to implement finite state automata and many other useful components, including cognitive maps and natural language parsers. These components support the development of, among other things, agents in virtual environments. Two spiking neuron agents are described, both able to run using either a standard neural simulator or using neuromorphic hardware. Examples of their behavior are described touching the individual spike level. The component model supports step-wise development, and the example of extending the cognitive mapping mechanism from the simple agent to the full agent is described. Spiking nets support parallelism, use on neuromorphic platforms, and engineering and exploration of multiple subsystems, which in turn can help explore the neural basis of cognitive phenomena. Relationships between these spiking nets and biology are discussed. The code is available to ease reuse by other researchers. Full article

On GPU-based clusters, the training workloads of machine learning (ML) models, particularly neural networks (NNs), are often structured as Directed Acyclic Graphs (DAGs) and typically deployed for parallel execution across heterogeneous GPU resources. Efficient scheduling of these workloads is crucial for optimizing performance metrics such as execution time, under various constraints including GPU heterogeneity, network capacity, and data dependencies. DAG-structured ML workload scheduling could be modeled as a Nonlinear Integer Program (NIP) problem, and is shown to be NP-complete. By leveraging a positive correlation between Scheduling Plan Distance (SPD) and Finish Time Gap (FTG) identified through an empirical study, we propose to develop a Running Time Gap Strategy for scheduling based on Whale Optimization Algorithm (WOA) and Reinforcement Learning, referred to as WORL-RTGS. The proposed method integrates the global search capabilities of WOA with the adaptive decision-making of Double Deep Q-Networks (DDQN). Particularly, we derive a novel function to generate effective scheduling plans using DDQN, enhancing adaptability to complex DAG structures. Comprehensive evaluations on practical ML workload traces collected from Alibaba on simulated GPU-enabled platforms demonstrate that WORL-RTGS significantly improves WOA’s stability for DAG-structured ML workload scheduling and reduces completion time by up to 66.56% compared with five state-of-the-art scheduling algorithms. Full article

The present study examines the relationship between economic growth and the tax burden that is formed as a result of income taxes. The main goal is to verify whether there is a link between these research variables in the long run and if this is confirmed, to analyze the manner in which these processes interact. The research applies a range of econometric techniques, including stationary tests, pairwise Granger causality test, Johansen cointegration test, impulse functions, and variance decompositions in order to investigate causality in the short- and long-term. The study is based on 49 observations and covers four European Union (EU) member states (Bulgaria, Hungary, Romania, and Estonia), which continue to impose a proportional (flat) tax on personal and corporate income. The analysis relies on quarterly data for the period 2013Q1–2025Q1. The results obtained are quite heterogeneous, which can be explained by the significant differences in the tax policy pursued, as well as by a number of other features determining the growth of national economies. Full article

To accommodate the high penetration of intermittent renewable energy sources like wind and solar power into the grid, coal-fired units are required to operate with enhanced deep peak-shaving and variable load capabilities. This study develops a dynamic model of the boiler–turbine coordinated control system (BTCCS) for ultra-supercritical once-through boiler (OTB) coal-fired units operating under wet conditions. A mechanistic model framework is established based on mass and energy conservation. In case of missing steady-state data, this work proposes a mechanism-integrated parameter identification method that determines model parameters using only dynamic running data while incorporating physical constraints. Model validation demonstrates that the proposed approach accurately reproduces the variable-load operation of the BTCCS within the range of 50–350 MW. Mean relative errors of output variables are all less than 7.5%, and root mean square errors of output variables are less than 0.3 MPa, 1.4 kg/s, 0.25 m, and 20.7 MW, respectively. Open-loop simulations further confirm that the model captures the essential dynamic characteristics of the system, making it suitable for simulation studies and control system design aimed at improving operational flexibility and safety of OTB coal-fired units under wet conditions. Full article

After the powerful socioeconomic shock of the fall of the communist regime in the early 90s, the ten countries in Central and Eastern Europe (CEE) analyzed in this study became growing Member States of the European Union (EU). However, they faced the 2008 financial crisis, the 2019 COVID shock, and sharp income disparities both at the regional level and compared to the countries in Western EU. This study explores the differences in sustainable regional development, modeling with Panel Autoregressive Distributed Lag (ARDL) to analyze relationships across multiple cross-sections in the short and long run, as well as with Cointegration Tests and Granger Panel Causality to detect evidence of causality among the variables in the study. The analysis covers 2012–2022, a period in which the Member States from CEE had the best access to generous structural and cohesion EU funds and that includes both the post-financial crisis convergence phase and the COVID-19 shock, enabling us to capture regional resilience dynamics. The results indicate that capital formation and population density positively influence disposable household income in the long run, across CEE regions, while unemployment and life expectancy exert negative effects. The results of this paper can be of use to decision-making institutions seeking to implement proactive socioeconomic policies in the lagging regions, before the next crisis, focused on capital investments, reducing unemployment, and bridging the rural–urban divide. The study contributes to the literature on inclusive and sustainable economic development at the CEE regional level. Full article

This paper analyzes a finite-capacity $GI/M/2/N$ queue with two heterogeneous servers operating under a multiple working-vacation policy, Bernoulli feedback, and customer impatience. Using the supplementary-variable technique in tandem with a tailored recursive scheme, we derive the stationary distributions of the system size as observed at pre-arrival instants and at arbitrary epochs. From these, we obtain explicit expressions for key performance metrics, including blocking probability, average reneging rate, mean queue length, mean sojourn time, throughput, and server utilizations. We then embed these metrics in an economic cost function and determine service-rate settings that minimize the total expected cost via the Bat Algorithm. Numerical experiments implemented in R validate the analysis and quantify the managerial impact of the vacation, feedback, and impatience parameters through sensitivity studies. The framework accommodates general renewal arrivals ($GI$), thereby extending classical ($M/M/2/N$) results to more realistic input processes while preserving computational tractability. Beyond methodological interest, the results yield actionable design guidance: (i) they separate Palm and time-stationary viewpoints cleanly under non-Poisson input, (ii) they retain heterogeneity throughout all formulas, and (iii) they provide a cost–optimization pipeline that can be deployed with routine numerical effort. Methodologically, we (i) characterize the generator of the augmented piecewise–deterministic Markov process and prove the existence/uniqueness of the stationary law on the finite state space, (ii) derive an explicit Palm–time conversion formula valid for non-Poisson input, (iii) show that the boundary-value recursion for the Laplace–Stieltjes transforms runs in linear time $\mathcal{O}\left(N\right)$ and is numerically stable, and (iv) provide influence-function (IPA) sensitivities of performance metrics with respect to $({\mu }_1,{\mu }_2,\nu ,\alpha ,\varphi ,\beta )$. Full article

This work presents a time-domain approach for characterizing the Ground Reaction Forces (GRFs) exerted by a pedestrian during running. It is focused on the vertical component, but the methodology is adaptable to other components or activities. The approach is developed from a statistical perspective. It relies on experimentally measured force-time series obtained from a healthy male pedestrian at eight step frequencies ranging from 130 to 200 steps/min. These data are subsequently used to build a stochastic data-driven model. The model is composed of multivariate normal distributions which represent the step patterns of each foot independently, capturing potential disparities between them. Additional univariate normal distributions represent the step scaling and the aerial phase, the latter with both feet off the ground. A dimensionality reduction procedure is also implemented to retain the essential geometric features of the steps using a sufficient set of random variables. This approach accounts for the intrinsic variability of running gait by assuming normality in the variables, validated through state-of-the-art statistical tests (Henze-Zirkler and Shapiro-Wilk) and the Box-Cox transformation. It enables the generation of virtual GRFs using pseudo-random numbers from the normal distributions. Results demonstrate strong agreement between virtual and experimental data. The virtual time signals reproduce the stochastic behavior, and their frequency content is also captured with deviations below 4.5%, most of them below 2%. This confirms that the method effectively models the inherent stochastic nature of running human gait. Full article

Napping is recognized as a strategy to enhance athletic performance. However, the optimal timing and duration for maximizing its benefits remain unclear. This study investigated the effects of a 90 min nap at different times on physical performance, psycho-cognitive responses, and perceived recovery in trained youth male athletes. Fourteen athletes (18 ± 1 years) completed four conditions in a randomized crossover design: (1) No-nap-13h, (2) No-nap-15h, (3) Nap-13h, and (4) Nap-15h. After each condition, athletes performed a 5 m shuttle run test (5mSRT) and were assessed on best distance (BD), total distance (TD), and fatigue index (FI). Ratings of perceived exertion (RPE) were recorded after each 5mSRT repetition, whereas muscle soreness (DOMS) and recovery (PRS) were assessed post-test and 24 h later. The digit cancelation test (DCT), feeling scale (FS), Stanford Sleepiness Scale (SSS), and Hooper Questionnaire evaluated sleep quality and psycho-cognitive state. Results showed that the athletes felt greater sleepiness before Nap-15h and after Nap-13h versus the no-nap conditions. TD was higher in Nap-13h than Nap-15h (p = 0.001) and No-nap-15h (p = 0.0009). BD was higher in Nap-13h versus No-nap-15h and No-nap-13h, while RPE was higher in Nap-13h versus No-nap-13 h, Nap-15h, and No-nap-15h (all, p < 0.05). DCT scores were also higher in Nap-13h. No significant effects were found for FI, FS, or Hooper. In conclusion, a 90 min nap at 13:00 was more effective than a later nap or no nap in improving performance and recovery, suggesting benefits for afternoon training or competitions. Full article

Hyperspectral imagers on satellites obtain the fine spectral signatures that are essential in distinguishing one material from another but at the expense of a limited spatial resolution. Enhancing the latter is thus a desirable preprocessing step in order to further improve the detection capabilities offered by hyperspectral images for downstream tasks. At the same time, there is growing interest in deploying inference methods directly onboard satellites, which calls for lightweight image super-resolution methods that can be run on the payload in real time. In this paper, we present a novel neural network design, called Deep Pushbroom Super-Resolution (DPSR), which matches the pushbroom acquisition of hyperspectral sensors by processing an image line by line in the along-track direction with a causal memory mechanism to exploit previously acquired lines. This design greatly limits the memory requirements and computational complexity, achieving onboard real-time performance, i.e., the ability to super-resolve a line in the time that it takes to acquire the next one, on low-power hardware. Experiments show that the quality of the super-resolved images is competitive with or even surpasses that of state-of-the-art methods that are significantly more complex. Full article

Temperature excursions during refrigerated transport strongly affect the quality and shelf life of perishable food, yet reproducing realistic, time-varying cold-chain temperature histories in the laboratory remains challenging. In this study, we present a compact, portable climate chamber driven by Peltier modules and an identification-guided control architecture designed to reproduce real refrigerated-truck temperature histories with high fidelity. Control is implemented as a cascaded regulator: an outer two-degree-of-freedom PID for air-temperature tracking and faster inner PID loops for module-face regulation, enhanced with derivative filtering, anti-windup back-calculation, a Smith predictor, and hysteresis-based bumpless switching to manage dead time and polarity reversals. The system integrates distributed temperature and humidity sensors to provide real-time feedback for precise thermal control, enabling accurate reproduction of cold-chain conditions. Validation comprised two independent 36-day reproductions of field traces and a focused 24-h comparison against traditional control baselines. Over the long trials, the chamber achieved very low long-run errors ($\mathrm{MAE}\cong 0.19 °\mathrm{C}$, $\mathrm{MedAE}\cong 0.10 °\mathrm{C}$, $\mathrm{RMSE}\cong 0.33 °\mathrm{C}$, $R^2=0.9985$). The 24-h test demonstrated that our optimized controller tracked the reference, improving both transient and steady-state behaviour. The system tolerated realistic humidity transients without loss of closed-loop performance. This portable platform functions as a reproducible physical twin for cold-chain experiments and a reliable data source for training predictive shelf-life and digital-twin models to reduce food waste. Full article

We analyze how to perform accurate time series forecasting for monitoring poultry health in a complex pen environment. To this end, we make use of a novel dataset consisting of a collection of real-world sensor data in the housing of turkeys. The dataset comprises features such as food intake, water intake, and various environmental values, which come with high variance, sensor defects, and unreliable timestamps. In this paper, we investigate different state-of-the-art forecasting algorithms to predict different features, as well as a variety of deep learning models such as different transformer models and time series foundational models. We evaluate both their forecasting accuracy as well as the efforts required to run the models in the first place. Our findings show that some of these aforementioned algorithms are able to produce satisfactory forecasting results on this highly challenging dataset while still remaining easy to use, which is key in a tech-distant industry such as poultry farming. Full article

Monitoring electrocardiogram (ECG) and respiration continuously and non-invasively is essential for managing cardiopulmonary health. An effective wearable device can be used to regularly monitor key vitals, reducing the need for clinical visits. In this work, we propose a custom device for real-time continuous ECG by inkjet printed (IJP) dry electrodes and respiration monitoring by using a novel single 6-axis inertial measurement unit (IMU). The proposed system can extract the heart rate (HR) and respiration rate (RR) during static and dynamic postures. The respiration process implements a quaternion-based update and multiple filtering stages to estimate the signal. The custom device uses Bluetooth protocol to send the raw and processed data to a mobile application. The RR is investigated in stationary, i.e., sitting and standing, and dynamic, i.e., walking, running, and cycling, postures. The proposed device is evaluated with commercial Go Direct^® respiration belt from Vernier^® for RR and offers an overall accuracy of 99.3% and 98.6% for static and dynamic conditions, respectively. The wearable also offers 98.9% and 97.9% accuracy for HR measurements, respectively, in static and active postures when compared with the Kardia^® device. Furthermore, the device is assessed in an ambulatory monitoring setup in both indoor and outdoor environments. The low-power wearable consumes an average of only 7.4 mA of current during data processing. The device performs effectively and efficiently in both stationary and active states, offering a low complexity, portable solution for real-time monitoring. The proposed system can benefit from the continuous monitoring and early detection of pulmonary and cardio-respiratory health issues. Full article

With the rapid development of autonomous vehicles and intelligent transportation systems, vehicle detection and distance estimation have become critical technologies for ensuring driving safety. However, real-world in-vehicle environments impose strict constraints on computational resources, making it impractical to deploy high-end GPUs. This implies that even highly accurate algorithms, if unable to run in real time on embedded platforms, cannot fully meet practical application demands. Although existing deep learning-based detection and stereo vision methods achieve state-of-the-art accuracy on public datasets, they often rely heavily on massive computational power and large-scale annotated data. Their high computational requirements and limited cross-scenario generalization capabilities restrict their feasibility in real-time vehicle-mounted applications. On the other hand, traditional algorithms such as Semi-Global Block Matching (SGBM) are advantageous in terms of computational efficiency and cross-scenario adaptability, but when used alone, their accuracy and robustness remain insufficient for safety-critical applications. Therefore, the motivation of this study is to develop a stereo vision-based collision warning system that achieves robustness, real-time performance, and computational efficiency. Our method is specifically designed for resource-constrained in-vehicle platforms, integrating a lightweight YOLOv8n detector with SGBM-based depth estimation. This approach enables real-time performance under limited resources, providing a more practical solution compared to conventional deep learning models and offering strong potential for real-world engineering applications. Full article

Stomach ulcers, a common type of gastrointestinal (GI) disease, pose serious health risks if not diagnosed and treated at an early stage. Therefore, in this research, a method is proposed based on two deep learning models for classification and segmentation. The classification model is based on Convolutional Neural Networks (CNN) and incorporates Q-learning to achieve learning stability and decision accuracy through reinforcement-based feedback. In this model, input images are passed through a custom CNN model comprising seven layers, including convolutional, ReLU, max pooling, flattening, and fully connected layers, for feature extraction. Furthermore, the agent selects an action (class) for each input and receives a +1 reward for a correct prediction and −1 for an incorrect one. The Q-table stores a mapping between image features (states) and class predictions (actions), and is updated at each step based on the reward using the Q-learning update rule. This process runs over 1000 episodes and utilizes Q-learning parameters (α = 0.1, γ = 0.6, ϵ = 0.1) to help the agent learn an optimal classification strategy. After training, the agent is evaluated on the test data using only its learned policy. The classified ulcer images are passed to the proposed attention-based U-Net model to segment the lesion regions. The model contains an encoder, a decoder, and attention layers. The encoder block extracts features through pooling and convolution layers, while the decoder block up-samples the features and reconstructs the segmentation map. Similarly, the attention block is used to highlight the important features obtained from the encoder block before passing them to the decoder block, helping the model focus on relevant spatial information. The model is trained using the selected hyperparameters, including an 8-batch size, the Adam optimizer, and 50 epochs. The performance of the models is evaluated on Kvasir, Nerthus, CVC-ClinicDB, and a private POF dataset. The classification framework provides 99.08% accuracy on Kvasir and 100% accuracy on Nerthus. In contrast, the segmentation framework yields 98.09% accuracy on Kvasir, 99.77% accuracy on Nerthus, 98.49% accuracy on CVC-ClinicDB, and 99.13% accuracy on the private dataset. The achieved results are superior to those of previous methods published in this domain. Full article

Neural networks are powerful, high-accuracy systems whose trained parameters represent a valuable intellectual property. Building models that reach top level performance is a complex task and requires substantial investments of time and money so protecting these assets is an increasingly important task. Extensive research has been carried out on Neural Network Watermarking, exploring the possibility of inserting a recognizable marker in a host model either in the form of a concealed bit-string or as a characteristic output, making it possible to confirm network ownership even in the presence of malicious attempts at erasing the embedded marker from the model. The study examines the applicability of Opt-FreeMark, a non-invasive post-training white-box watermarking technique, obtained by modifying and optimizing an already existing state-of-the-art technique for tiny neural networks. Here, “Tiny” refers to models intended for ultra-low-power deployments, such as those running on edge devices like sensors and micro-controllers. Watermark robustness is also demonstrated by simulating common model-modification attacks that try to eliminate it from the model while preserving performance; the results presented in the paper indicate that the watermarking scheme effectively protects the networks against these manipulations. Full article