Search Results (169)

The optimization of high-frequency circuits remains a computationally intensive task due to the need for repeated high-fidelity electromagnetic (EM) simulations. To address this challenge, we propose a novel integration of machine learning-generated inverse maps within the space mapping (SM) optimization framework to significantly accelerate circuit optimization while maintaining high accuracy. The proposed approach leverages Bayesian Neural Networks (BNNs) and surrogate modeling techniques to construct an inverse mapping function that directly predicts design parameters from target performance metrics, bypassing iterative forward simulations. The methodology was validated using a low-pass filter optimization scenario, where the inverse surrogate model was trained using electromagnetic simulations from COMSOL Multiphysics 2024 r6.3 and optimized using MATLAB R2024b r24.2 trust region algorithm. Experimental results demonstrate that our approach reduces the number of high-fidelity simulations by over 80% compared to conventional SM techniques while achieving high accuracy with a mean absolute error (MAE) of 0.0262 (0.47%). Additionally, convergence efficiency was significantly improved, with the inverse surrogate model requiring only 31 coarse model simulations, compared to 580 in traditional SM. These findings demonstrate that machine learning-driven inverse surrogate modeling significantly reduces computational overhead, accelerates optimization, and enhances the accuracy of high-frequency circuit design. This approach offers a promising alternative to traditional SM methods, paving the way for more efficient RF and microwave circuit design workflows. Full article

Effective earthquake early warning (EEW) is essential for disaster prevention in the built environment, enabling a rapid structural response, system shutdown, and occupant evacuation to mitigate damage and casualties. However, most current EEW systems lack rigorous reliability analyses of their predictive outcomes, limiting their effectiveness in real-world scenarios—especially for on-site warnings, where data are limited and time is critical. To address these challenges, we propose a Bayesian neural network (BNN) framework based on Stein variational gradient descent (SVGD). By performing Bayesian inference, we estimate the posterior distribution of the parameters, thus outputting a reliability analysis of the prediction results. In addition, we incorporate a continual learning mechanism based on elastic weight consolidation, allowing the system to adapt quickly without full retraining. Our experiments demonstrate that our SVGD-BNN model significantly outperforms traditional peak displacement (Pd)-based approaches. In a 3 s time window, the Pearson correlation coefficient R increases by 9.2% and the residual standard deviation $SD$ decreases by 24.4% compared to a variational inference (VI)-based BNN. Furthermore, the prediction variance generated by the model can effectively reflect the uncertainty of the prediction results. The continual learning strategy reduces the training time by 133–194 s, enhancing the system’s responsiveness. These features make the proposed framework a promising tool for real-time, reliable, and adaptive EEW—supporting disaster-resilient building design and operation. Full article

This paper presents a dynamic model for full-power converter permanent magnet synchronous wind turbines based on Physics-Informed Neural Networks (PINNs). The model integrates the physical dynamics of the wind turbine directly into the loss function, enabling high-accuracy equivalent modeling with limited data and overcoming the typical “black-box” constraints and large data requirements of traditional data-driven approaches. To enhance the model’s real-time adaptability, we introduce an online update mechanism leveraging Bayesian Neural Networks (BNNs) combined with a clustering-guided strategy. This mechanism estimates uncertainty in the neural network weights in real-time, accurately identifies error sources, and performs local fine-tuning on clustered data. This improves the model’s ability to track real-time errors and addresses the challenge of parameter-specific adjustments. Finally, the data-driven model is integrated into the CloudPSS platform, and its multi-scenario modeling accuracy is validated across various typical cases, demonstrating the robustness of the proposed approach. Full article

Modern traffic management requires sophisticated approaches to address the complexities of urban road networks, which continue to grow in complexity due to increasing urbanization and vehicle usage. Traditional methods often fall short in mitigating congestion and optimizing traffic flow, inducing the exploration of innovative traffic control strategies based on advanced theoretical frameworks. In this sense, we explore different game theory-based control strategies in an eight-intersection traffic network modeled by means of hybrid systems and graph theory, using a software simulator that combines the multi-modal traffic simulation software VISSIM and MATLAB to integrate traffic network parameters and population game criteria. Across five distinct network scenarios with varying saturation conditions, we explore a fixed-time scheme of signaling by means of fictitious play dynamics and adaptive schemes, using dynamics such as Smith, replicator, Logit and Brown–Von Neumann–Nash (BNN). Results show better performance for Smith and replicator dynamics in terms of traffic parameters both for fixed and variable signaling times, with an interesting outcome of fictitious play over BNN and Logit. Full article

This study projects future Residential Water Use (RWU) under climate change scenarios using a Bayesian Neural Network (BNN) model that quantifies the relationship between observed temperatures and RWU. Eighteen Global Climate Models (GCMs) under the Shared Socioeconomic Pathway 5–8.5 (SSP5–8.5) scenario were used to assess the uncertainties across these models. The findings indicate that RWU in Republic of Korea (ROK) is closely linked to temperature changes, with significant increases projected in the distant future (F3), especially during summer. Under the SSP5–8.5 scenario, RWU is expected to increase by up to 10.3% by the late 21st century (2081–2100) compared to the historical baseline. The model achieved a root mean square error (RMSE) of 11,400 m³/month, demonstrating reliable predictive performance. Unlike conventional deep learning models, the BNN provides probabilistic forecasts with uncertainty bounds, enhancing its suitability for climate-sensitive resource planning. This study also projects inflows to the Paldang Dam, revealing an overall increase in future water availability. However, winter water security may decline due to decreased inflow and minimal changes in RWU. This study suggests enhancing summer precipitation storage while considering downstream flood risks. Demand management strategies are recommended for addressing future winter water security challenges. This research highlights the importance of projecting RWU under climate change scenarios and emphasizes the need for strategic water resource management in ROK. Full article

For data-intensive applications like edge AI and image processing, we present a new reconfigurable 8T SRAM-based in-memory computing (IMC) macro designed for high-performance and energy-efficient operation. This architecture mitigates von Neumann limitations through numerous major breakthroughs. We built a new architecture with an adjustable capacitance array to substantially increase the multiply-and-accumulate (MAC) engine’s accuracy. It achieves 10–20 TOPS/W and >95% accuracy for 4–10-bit operations and is robust across PVT changes. By supporting binary and ternary neural networks (BNN/TNN) with XNOR-and-accumulate logic, a dual-mode inference engine further expands capabilities. With sub-5 ns mode switching, it can achieve up to 30 TOPS/W efficiency and >97% accuracy. In-memory Hamming error correction is implemented directly using integrated XOR circuitry. This technique eliminates off-chip ECC with >99% error correction and >98% MAC accuracy. Machine learning-aided co-optimization ensures sense amplifier dependability. To ensure CMOS compatibility, the macro may perform Boolean logic operations using normal 8T SRAM cells. Comparative circuit-level simulations show a 31.54% energy efficiency boost and a 74.81% delay reduction over other SRAM-based IMC solutions. These improvements make our macro ideal for real-time AI acceleration, cryptography, and next-generation edge computing, enabling advanced compute-in-memory systems. Full article

With the rapid development of marine renewable energy, especially offshore photovoltaic systems, the problem of biofouling of photovoltaic equipment in the marine environment has become increasingly prominent. The attachment of marine organisms such as algae will significantly affect the photoelectric conversion efficiency of photovoltaic panels, thereby reducing the stability and economy of the system. In this study, a composite siloxane coating was designed and prepared. Methyltrimethoxysilane (MTMS) was used as the organosilicon component. The negative potential of the coating was significantly enhanced by incorporating hexagonal boron nitride nanosheets (h-BNNS). This negative potential and the negative charge on the surface of marine organisms, especially algae, would produce electrostatic repulsion, which can effectively reduce the attachment of organisms. The results show that the prepared coating exhibits excellent performance in anti-biofouling, adhesion, chemical stability, transparency, and self-cleaning properties. The transparency of the coating reached 92.7%. After immersion with Chlorella for 28 days, the coverage percentage on the coating surface was only 0.98%, while the coverage percentage on the blank sample was 23.25%. The corrosion resistance and salt resistance of the coating also ensure its stability in complex marine environments, and it has broad application prospects. Full article

A significant advancement in Neural Network (NN) research is the integration of domain-specific knowledge through custom loss functions. This approach addresses a crucial challenge: How can models utilize physics or mathematical principles to enhance predictions when dealing with sparse, noisy, or incomplete data? Physics-Informed Neural Networks (PINNs) put this idea into practice by incorporating a forward model, such as Partial Differential Equations (PDEs), as soft constraints. This guidance helps the networks find solutions that align with established laws. Recently, researchers have expanded this framework to include Bayesian NNs (BNNs) which allow for uncertainty quantification. However, what happens when the governing equations of a system are not completely known? In this work, we introduce methods to automatically select PDEs from historical data in a parametric family. We then integrate these learned equations into three different modeling approaches: PINNs, Bayesian-PINNs (B-PINNs), and Physical-Informed Bayesian Linear Regression (PI-BLR). To assess these frameworks, we evaluate them on a real-world Multivariate Time Series (MTS) dataset related to electrical power energy management. We compare their effectiveness in forecasting future states under different scenarios: with and without PDE constraints and accuracy considerations. This research aims to bridge the gap between data-driven discovery and physics-guided learning, providing valuable insights for practical applications. Full article

Broad lateral size and thickness distributions impede the application of hexagonal boron nitride nanosheets (BNNSs) as friction modifiers in base oil, although they possess remarkable potential for lubrication performance promotion. In this work, a cascade centrifugation-assisted liquid-phase exfoliation approach was presented to prepare BNNSs from hexagonal boron nitride (h-BN) efficiently and scalably. Subsequently, they were ultrasonically dispersed into gas-to-liquid (GTL) base oil, and their lubrication performance promotion was evaluated by a four-ball tribotester. Tribological tests demonstrated that BNNS possesses excellent friction-reducing and anti-wear properties in GTL. Furthermore, the findings indicate that at a BNNS content of 0.8 wt.%, the system displayed the lowest COF and WSD. Particularly, with an addition of 0.8 wt.% BNNS into GTL, the AFC and WSD are reduced significantly by 40.1% and 35.4% compared to pure base oil, respectively, and the surface roughness, wear depth, and wear volume were effectively reduced by 91.0%, 68.5%, and 76.8% compared to GTL base oil, respectively. Raman, SEM-EDS, and XPS results proved that the outstanding friction-reducing and anti-wear properties of BNNS can mainly be ascribed to the presence of physical adsorption film and tribo-chemical film, which were composed of FeOOH, FeO, Fe₃O₄, and B₂O₃. Full article

Batch normalization (BN) is crucial for achieving state-of-the-art binary neural networks (BNNs). Unlike full-precision neural networks, BNNs restrict activations to discrete values $\{-1,1\}$, which requires a renewed understanding and research of the role and significance of the BN layers in BNNs. Many studies notice this phenomenon and try to explain it. Inspired by these studies, we introduce the self-attention mechanism into BN and propose a novel Attention-Based Batch Normalization (ABN) for Binary Neural Networks. Also, we present an ablation study of parameter trade-offs in ABN, as well as an experimental analysis of the effect of ABN on BNNs. Experimental analyses show that our ABN method helps to capture image features, provide additional activation-like functions, and increase the imbalance of the activation distribution, and these features help to improve the performance of BNNs. Furthermore, we conduct image classification experiments over the CIFAR10, CIFAR100, and TinyImageNet datasets using BinaryNet and ResNet-18 network structures. The experimental results demonstrate that our ABN consistently outperforms the baseline BN across various benchmark datasets and models in terms of image classification accuracy. In addition, ABN exhibits less variance on the CIFAR datasets, which suggests that ABN can improve the stability and reliability of models. Full article

During tissue injury or infection, leukocytes are activated to produce proinflammatory mediators, which trigger the immune system to produce anti-inflammatory and analgesic molecules. Our previous studies provide evidence that synthetic microneurotrophins, like BNN27, exert significant analgesic and anti-inflammatory effects during Complete Freund’s Adjuvant (CFA)-induced inflammation and pain. Thus, the aim of the present study was to examine if the effect of BNN27 on inflammatory pain is mediated at least in part by activation of T-lymphocytes. For this purpose, six hours following the injection of CFA, spleens were harvested in PBS and lymphocytes were collected and placed in medium containing concanavalin-A and IL-2 to prompt T-lymphocyte proliferation and differentiation. Cells were then treated with BNN27 at different concentrations and the media and cells were collected for ELISA and PCR assays. The proliferation rate of T-cells was also examined using the MTT assay. Our results showed that BNN27 significantly increased the proliferation of T-lymphocytes. In addition, BNN27 significantly decreased IL-6 and TNF-α protein levels, while it increased the mRNA expression of μ-opioid receptor and opioid peptides PENK and POMC at different time points. Our data demonstrate considerable anti-inflammatory and analgesic effects of BNN27, making it a promising molecule for inflammation and pain management. Full article

The integration of in vitro biological neural networks (BNNs) with robotic systems to explore their information processing and adaptive learning in practical tasks has gained significant attention in the fields of neuroscience and robotics. However, existing BNN-based robotic systems cannot perceive the visual environment due to the inefficiency of sensory information encoding methods. In this study, we propose a biomimetic visual information spatiotemporal encoding method based on improved delayed phase encoding. This method transforms high-dimensional images into a series of pulse sequences through convolution, temporal delay, alignment, and compression for BNN stimuli. We conduct three stages of unsupervised training on in vitro BNNs using high-density microelectrode arrays (HD-MEAs) to validate the potential of the proposed encoding method for image recognition tasks. The neural activity is decoded via a logistic regression model. The experimental results show that the firing patterns of BNNs with different spatiotemporal stimuli are highly separable in the feature space. After the third training stage, the image recognition accuracy reaches 80.33% ± 7.94%, which is 13.64% higher than that of the first training stage. Meanwhile, the BNNs exhibit significant increases in the connection number, connection strength, and inter-module participation coefficient after unsupervised training. These results demonstrate that the proposed method significantly enhances the functional connectivity and cross-module information exchange in BNNs. Full article

Parking availability prediction is a critical component of intelligent transportation systems, aiming to reduce congestion and improve urban mobility. While traditional deep learning models such as Long Short-Term Memory (LSTM) networks have been widely applied, they lack mechanisms to quantify uncertainty, limiting their robustness in real-world deployments. This paper proposes a Bayesian Neural Network (BNN)-based framework for parking occupancy prediction that explicitly models both epistemic and aleatoric uncertainty. Although BNNs have shown promise in other domains, they remain underutilised in parking prediction—likely due to the computational complexity and the absence of real-time context integration in earlier approaches. Our approach leverages contextual features, including temporal and environmental factors, to enhance uncertainty-aware predictions. The framework is evaluated under varying data conditions, including data scarcity (90%, 50%, and 10% of training data) and synthetic noise injection to simulate aleatoric uncertainty. Results demonstrate that BNNs outperform other methods, achieving an average accuracy improvement of 27.4% in baseline conditions, with consistent gains under limited and noisy data. Applying uncertainty thresholds at 20% and 30% further improves reliability by enabling selective, confidence-based decision making. This research shows that modelling both types of uncertainty leads to significantly improved predictive performance in intelligent transportation systems and highlights the potential of uncertainty-aware approaches as a foundation for future work on integrating BNNs with hybrid neuro-symbolic reasoning to enhance decision making under uncertainty. Full article

This work introduces the use of Bayesian Neural Networks (BNNs) for inferring chlorophyll-a concentration ([CHL-a]) from remotely sensed data. BNNs are probabilistic models that associate a probability distribution to the neural network parameters and rely on Bayes’ rule for training. The performance of the proposed probabilistic model is compared to that of standard ocean color algorithms, namely ocean color 4 (OC4) and ocean color index (OCI). An extensive in situ bio-optical dataset was used to train and validate the ocean color models. In contrast to established methods, the BNN allows for enhanced modeling flexibility, where different variables that affect phytoplankton phenology or describe the state of the ocean can be used as additional input for enhanced performance. Our results suggest that BNNs perform at least as well as established methods, and they could achieve 20–40% lower mean squared errors when additional input variables are included, such as the sea surface temperature and its climatological mean alongside the coordinates of the prediction. The BNNs offer means for uncertainty quantification by estimating the probability distribution of [CHL-a], building confidence in the [CHL-a] predictions through the variance of the predictions. Furthermore, the output probability distribution can be used for risk assessment and decision making through analyzing the quantiles and shape of the predicted distribution. Full article

Background: Bone elasticity is one of the most important biomechanical parameters of the skeleton. It varies markedly with age, anatomical zone, bone type (cortical or trabecular) and bone marrow status. Methods: This review presents the result of a systematic review and analysis of 495 experimental and analytical papers on the elastic properties of bone tissue. The bone characteristics of hip, shoulder, skull, vertebrae as a function of the factors of age (young and old), sex (male and female), presence/absence of bone marrow and different test methods are examined. The Bayesian neural network (BNN) was used to estimate the uncertainty in some skeletal parameters (age, sex, and body mass index) in predicting bone elastic modulus. Results: It was found that the modulus of elasticity of cortical bone in young people is in the range of 10–30 GPa (depending on the type of bone), and with increasing age, this slightly decreases to 10–25 GPa, while trabecular tissue varies from 0.2 to 5 GPa and reacts more acutely to osteoporosis. Bone marrow, according to several studies, is able to partially increase stiffness under impact loading, but its contribution is minimal under slow deformations. Conclusions: BNN confirmed high variability, supplementing the predictions with confidence intervals and allowed the formation of equations for the calculation of bone tissue elastic modulus for the subsequent selection of the recommended elastic modulus of the finished implant, taking into account the biomechanical characteristics of bone tissue depending on age (young and old), sex (men and women) and anatomical zones of the human skeleton. Full article