Search Results (795)

Surface deformation induced by the extraction of natural resources constitutes a non-stationary spatiotemporal process. Modeling surface deformation time series obtained through Interferometric Synthetic Aperture Radar (InSAR) technology using deep learning methods is crucial for disaster prevention and mitigation. However, the complexity of model hyperparameter configuration and the lack of interpretability in the resulting predictions constrain its engineering applications. To enhance the reliability of model outputs and their decision-making value for engineering applications, this study presents a workflow that combines a Tree-structured Parzen Estimator (TPE)-based Bayesian optimization approach with ensemble inference. Using the Rhineland coalfield in Germany as a case study, we systematically evaluated six deep learning architectures in conjunction with various spatiotemporal coding strategies. Pairwise comparisons were conducted using a Welch t-test to evaluate the performance differences across each architecture under two parameter-tuning approaches. The Benjamini–Hochberg method was applied to control the false discovery rate (FDR) at 0.05 for multiple comparisons. The results indicate that TPE-optimized models demonstrate significantly improved performance compared to their manually tuned counterparts, with the ResNet+Transformer architecture yielding the most favorable outcomes. A comprehensive analysis of the spatial residuals further revealed that TPE optimization not only enhances average accuracy, but also mitigates the model’s prediction bias in fault zones and mineralize areas by improving the spatial distribution structure of errors. Based on this optimal architecture, we combined the ten highest-performing models from the optimization stage to generate a quantile-based susceptibility map, using the ensemble median as the central predictor. Uncertainty was quantified from three complementary perspectives: ensemble spread, class ambiguity, and classification confidence. Our analysis revealed spatial collinearity between physical uncertainty and absolute residuals. This suggests that uncertainty is more closely related to the physical complexity of geological discontinuities and human-disturbed zones, rather than statistical noise. In the analysis of super-threshold probability, the threshold sensitivity exhibited by the mining area reflects the widespread yet moderate impact of mining activities. By contrast, the fault zone continues to exhibit distinct high-probability zones, even under extreme thresholds. It suggests that fault-controlled deformation is more physically intense and poses a greater risk of disaster than mining activities. Finally, we propose an engineering decision strategy that combines uncertainty and residual spatial patterns. This approach transforms statistical diagnostics into actionable, tiered control measures, thereby increasing the practical value of susceptibility mapping in the planning of natural resource extraction. Full article

With the intensification of global climate change, frequent extreme temperature events pose increasing challenges to agricultural production. The aim of this study is to characterize the spatiotemporal evolution of county-level agricultural total factor productivity (ATFP) under extreme temperature events, reveal key driving factors and crop-specific heterogeneity, and predict potential high-risk areas, which is crucial for providing scientific basis for risk management and adaptive policy formulation in globally climate-sensitive agricultural regions. This paper selects Jiangsu Province as a typical case study, uses the DEA-Malmquist model to measure agricultural total factor productivity (ATFP), systematically analyzes the spatiotemporal dynamic evolution characteristics of ATFP at the county scale, and selects the random forest and XGBoost ensemble models with optimal accuracy through model comparison for prediction, assessing the evolution trends of ATFP under different climate scenarios. The results showed that: (1) From 1993 to 2022, the average ATFP increased from 0.7460 to 1.1063 in the province, though development showed uneven distribution across counties, exhibiting a “high in the south, low in the north” gradient pattern. (2) Mechanization, agricultural film and land inputs are the core elements driving the overall ATFP increase but there are obvious crop differences: mechanization has a more prominent role in promoting the productivity of wheat and maize, while labor inputs have a greater impact on the ATFP of rice. (3) The negative impacts of extreme climate events on agricultural production will be significantly amplified under high-emission scenarios, while moderate climate change may have a promotional effect on certain crops in some regions. Full article

The Arctic is warming rapidly, with atmospheric rivers (ARs) amplifying ice melt, extreme precipitation, and abrupt temperature shifts. Detecting ARs in the Arctic remains challenging, because AR detection algorithms designed for mid-latitudes perform poorly in polar regions. This study introduces a regional deep learning (DL) image segmentation model for Arctic AR detection, leveraging large-ensemble (LE) climate simulations. We analyse historical simulations from the Climate Change in the Arctic and North Atlantic Region and Impacts on the UK (CANARI) project, which provides a large, internally consistent sample of AR events at 6-hourly resolution and enables a close comparison of AR climatology across model and reanalysis data. A polar-specific, rule-based AR detection algorithm was adapted to label ARs in simulated data using multiple thresholds, providing training data for the segmentation model and supporting sensitivity analyses. U-Net-based models are trained on integrated water vapour transport, total column water vapour, and 850 hPa wind speed fields. We quantify how AR identification depends on threshold choices in the rule-based algorithm and show how these propagate to the U-Net-based models. This study represents the first use of the CANARI-LE for Arctic AR detection and introduces a unified framework combining rule-based and DL methods to evaluate model sensitivity and detection robustness. Our results demonstrate that DL segmentation achieves robust skill and eliminates the need for threshold tuning, providing a consistent and transferable framework for detecting Arctic ARs. This unified approach advances high-latitude moisture transport assessment and supports improved evaluation of Arctic extremes under climate change. Full article

We present LazyNet, a compact one-step neural-ODE model for single-cell CRISPR activation/interference (A/I) that operates directly on two-snapshot (“pre → post”) measurements and yields parameters with clear mechanistic meaning. The core log–linear–exp residual block exactly represents multiplicative effects, so synergistic multi-locus responses appear as explicit components rather than opaque composites. On a 53k-cell × 18k-gene neuronal Perturb-seq matrix, a three-replica LazyNet ensemble trained under a matched 1 h budget achieved strong threshold-free ranking and competitive error (genome-wide r ≈ 0.67) while running on CPUs. For comparison, we instantiated transformer (scGPT-style) and state-space (RetNet/CellFM-style) architectures from random initialization and trained them from scratch on the same dataset and within the same 1 h cap on a GPU platform, without any large-scale pretraining or external data. Under these strictly controlled, low-data conditions, LazyNet matched or exceeded their predictive performance while using far fewer parameters and resources. A T-cell screen included only for generalization showed the same ranking advantage under the identical evaluation pipeline. Beyond prediction, LazyNet exposes directed, local elasticities; averaging Jacobians across replicas produces a consensus interaction matrix from which compact subgraphs are extracted and evaluated at the module level. The resulting networks show coherent enrichment against authoritative resources (large-scale co-expression and curated functional associations) and concordance with orthogonal GPX4-knockout proteomes, recovering known ferroptosis regulators and nominating testable links in a lysosomal–mitochondrial–immune module. These results position LazyNet as a practical option for from-scratch, low-data CRISPR A/I studies where large-scale pretraining of foundation models is not feasible. Full article

The present study has applied a probabilistic oil spill modeling framework to assess the potential risks associated with offshore oil spills in the Foz do Amazonas sedimentary basin, a region of exceptional ecological importance and increasing geopolitical and socio-environmental relevance. By integrating a large ensemble of simulations with validated hydrodynamic, atmospheric and wave-driven forcings, the analysis of said simulations has provided a robust and seasonally resolved assessment of oil drift and beaching patterns along the Guianas and the Brazilian Equatorial Margin. The model has presented a total of 47,500 simulations performed on 95 drilling sites located across the basin, using the Lagrangian Spill, Transport and Fate Model (STFM) and incorporating a six-year oceanographic and meteorological variability. The simulations have included ocean current fields provided by HYCOM, wind forcing provided by GFS and Stokes drift provided by ERA5. Model performance has been evaluated by comparisons with satellite-tracked surface drifters using normalized cumulative Lagrangian separation metrics and skill scores. Mean skill scores have reached 0.98 after 5 days and 0.95 after 10 days, remaining above 0.85 up to 20 days, indicating high reliability for short to intermediate forecasting horizons and suitability for probabilistic applications. Probabilistic simulations have revealed a pronounced seasonal effect, governed by the annual migration of the Intertropical Convergence Zone (ITCZ). During the JFMA period, shoreline impact probabilities have exceeded 40–50% along extensive portions of the French Guiana and Amapá state (Brazil) coastlines, with oil reaching the coast typically within 10–20 days. In contrast, during the JASO period, beaching probabilities have decreased to below 15%, accompanied by a substantial reduction in impact along the coastline and higher variability in arrival times. Although coastal exposure has been markedly reduced during JASO, a residual probability of approximately 2% of oil intrusion into the Amazonas river mouth has persisted. Full article

This work presents a unified framework for multiobjective analog circuit optimization that combines surrogate modeling, uncertainty-aware evolutionary search, and adaptive high-fidelity verification. The approach integrates ensemble regressors and graph-based surrogate models with a closed-loop multi-fidelity controller that selectively invokes SPICE evaluations based on predictive uncertainty and diversity criteria. The framework includes reproducible caching, metadata tracking, and process- and Dask-based parallelism to reduce redundant simulations and improve throughput. The methodology is evaluated on four CMOS operational-amplifier topologies using NSGA-II, NSGA-III, SPEA2, and MOEA/D under a uniform configuration to ensure fair comparison. Surrogate-Guided Optimization (SGO) replaces approximately 96.5% of SPICE calls with fast model predictions, achieving about a 20× reduction in total simulation time while maintaining close agreement with ground-truth Pareto fronts. Multi-Fidelity Optimization (MFO) further improves robustness through adaptive verification, reducing SPICE usage by roughly 90%. The results show that the proposed workflow provides substantial computational savings with consistent Pareto-front quality across circuit families and algorithms. The framework is modular and extensible, enabling quantitative evaluation of analog circuits with significantly reduced simulation cost. Full article

Autonomous driving technology is advancing rapidly, particularly in vision-based approaches that use cameras to perceive the driving environment, which is the most human-like perception method. However, one of the key challenges that smart vehicles face is adapting to various weather conditions, which can significantly impact visual perception and vehicular control strategies. The ideal design for the latter is to dynamically adjust in real time to ensure safe and efficient driving, taking into account the prevailing weather conditions. In this study, we propose a lightweight weather perception model that incorporates multi-scale feature learning, channel attention mechanisms, and a soft voting ensemble strategy. This enables the model to capture various visual patterns, emphasize critical information, and integrate predictions across multiple modules for improved robustness. Benchmark comparisons are conducted using several well-known deep learning networks, including EfficientNet-B0, ResNet50, SqueezeNet, MobileNetV3-Large, MobileNetV3-Small, and LSKNet. Finally, using both public datasets and real-world video recordings from roads in Taiwan, our model demonstrates superior computational efficiency while maintaining high predictive accuracy. For example, our model achieves $98.07\%$ classification accuracy with only 0.4 million parameters and 0.19 GFLOPs, surpassing several well-known CNNs in computational efficiency. Compared with EfficientNet-B0, which has a similar accuracy ($98.37\%$) but requires over ten times more parameters and four times more FLOPs, our model offers a much lighter and faster alternative. Full article

Human beings are affected by different types of skin diseases worldwide. Automatic identification of skin disease from Dermoscopy images has proved effective for diagnosis and treatment to reduce fatality rate. The objective of this work is to demonstrate efficiency of three deep learning pre-trained models, namely MobileNet, EfficientNetB0, and DenseNet121 with ensembling techniques for classification of skin lesion images. This study considers HAM1000 dataset which consists of n = 10,015 images of seven different classes, with a huge class imbalance. The study has two-fold contributions for the classification methodology of skin lesions. First, modification of three pre-trained deep learning models for grouping of skin lesion into seven types. Second, Weighted Grid Search algorithm is proposed to address the class imbalance problem for improving the accuracy of the base classifiers. The results showed that the weighted ensembling method achieved a 3.67% average improvement in Accuracy, Precision, and Recall, 3.33% average improvement for F1-Score, and 7% average improvement for Matthews Correlation Coefficient (MCC) when compared to base classifiers. Evaluation of the model’s efficiency and performance shows that it obtained the highest ROC-AUC score of 92.5% for the modified MobileNet model for skin lesion categorization in comparison to EfficientNetB0 and DenseNet121, respectively. The implications of the results show that deep learning methods and classification techniques are effective for diagnosis and treatment of skin lesion diseases to reduce fatality rate or detect early warnings. Full article

Monitoring the battery state of health (SOH) has become increasingly important for electric vehicles (EVs), renewable storage systems, and consumer gadgets. It indicates the residual usable capacity and performance of a battery in relation to its original specifications. This information is crucial for the safety and performance enhancement of the overall system. This paper develops an explainable machine learning protocol with Bayesian optimization techniques trained on electrochemical impedance spectroscopy (EIS) data to predict battery SOH. Various robust ensemble algorithms, including HistGradientBoosting (HGB), Random Forest, AdaBoost, Extra Trees, Bagging, CatBoost, Decision Tree, LightGBM, Gradient Boost, and XGB, have been developed and fine-tuned for predicting battery health. Eight comprehensive metrics are employed to estimate the model’s performance rigorously: coefficient of determination (R²), mean squared error (MSE), median absolute error (medae), mean absolute error (MAE), correlation coefficient (R), Nash–Sutcliffe efficiency (NSE), Kling–Gupta efficiency (KGE), and root mean squared error (RMSE). Bayesian optimization techniques were developed to optimize hyperparameters across all models, ensuring optimal implementation of each algorithm. Feature importance analysis was performed to thoroughly evaluate the models and assess the features with the most influence on battery health degradation. The comparison indicated that the GradientBoosting model outperformed others, achieving an MAE of 0.1041 and an R² of 0.9996. The findings suggest that Bayesian-optimized tree-based ensemble methods, particularly gradient boosting, excel at forecasting battery health status from electrochemical impedance spectroscopy data. This result offers an excellent opportunity for practical use in battery management systems that employ diverse industrial state-of-health assessment techniques to enhance battery longevity, contributing to sustainability initiatives for second-life lithium-ion batteries. This capability enables the recycling of vehicle batteries for application in static storage systems, which is environmentally advantageous and ensures continuity. Full article

Magnetic levitation (Maglev) technology has long garnered significant attention in the engineering community due to its inherent advantages, such as contactless operation, minimal friction losses, low noise, and high precision. Based on electromagnetic suspension (EMS) and electrodynamic principles, these systems are primarily developed for advanced transportation, while also inspiring emerging applications such as vibration isolation and flywheel energy storage. Despite progress, practical deployment faces critical challenges, including accurate modeling, robustness against nonlinear and uncertain dynamics, and control stability under complex conditions. Artificial intelligence (AI), particularly machine learning (ML) offers promising solutions. Studies show ML-based methods, i.e., improved particle swarm optimization (PSO) optimize proportional-integral-derivative (PID) to reduce controller output overshoot, deep reinforcement learning (DRL) reduces levitation gap fluctuation under complex conditions, ensemble learning achieves high fault diagnosis accuracy, and convolutional neural network-long short-term memory (CNN-LSTM) predictive maintenance cuts costs. This review summarizes recent AI-enabled advances in Maglev transportation system modeling, control, and optimization, highlighting representative algorithms, performance comparisons, technical challenges, and future directions toward intelligent, reliable, and energy-efficient transportation systems. Full article

Student learning outcome (SLO) tracing aims to monitor students’ learning progress by predicting their likelihood of passing or failing courses using Deep Knowledge Tracing (DKT). However, conventional DKT models often lack interpretability, limiting their adoption in educational settings that require transparent decision-making. To address this challenge, this quantitative study proposes an interpretable ensemble framework that integrates Item Response Theory (IRT) with DKT. Specifically, multiple IRT-based DKT models are developed to capture student ability and item characteristics, and these models are combined using a bagging strategy to enhance predictive performance and robustness. The framework is evaluated on an SLO tracing dataset from Prairie View A&M University (PVAMU), a historically Black college and university (HBCU). Result analysis includes comparisons of evaluation metrics such as Area Under the Curve (AUC), accuracy (ACC), and precision across individual and ensemble models, as well as visualizations of student ability, item difficulty, and predicted probabilities to assess interpretability. Experimental results demonstrate that the ensemble approach consistently outperforms single models while providing clear, interpretable insights into student learning dynamics. These findings suggest that integrating ensemble methods with IRT can simultaneously improve prediction accuracy and transparency in SLO tracing. Full article

Background: Alzheimer’s disease (AD) is a leading worldwide cause of cognitive impairment, necessitating accurate, inexpensive diagnostic tools to enable early recognition. Methods: In this study, we present a robust deep learning approach for AD classification based on structural MRI scans, ConvNeXt, an emergent convolutional architecture inspired by vision transformers. We introduce AlzaSet, a clinically curated T1-weighted MRI dataset of 79 subjects (63 with Alzheimer’s disease [AD], 16 cognitively normal controls [NC]) acquired on a 1.5 T Siemens Aera in axial, coronal, and sagittal planes, respectively (12,947 slices in total). Images are neuroradiologist-labeled. Results are reported per plane, with awareness of the class imbalance at the subject level. We further present AlzaSet, a novel, expertly labeled clinical dataset with axial, coronal, and sagittal perspectives from AD and cognitively normal control subjects. Three ConvNeXt sizes (Tiny, Small, Base) were compared and benchmarked against existing state-of-the-art CNN models (VGG16, VGG19, InceptionV3, DenseNet121). Results: ConvNeXt-Base consistently outperformed the other models on coronal slices with an accuracy of 98.37% and an AUC of 0.992. Coronal views were determined to be most diagnostically informative, with emphasis on visualization of the medial temporal lobe. Moreover, comparison with recent ensemble-based techniques showed superior performance with comparable computational efficiency. Conclusions: These results indicate that ConvNeXt-capable models applied to clinically curated datasets have strong potential to provide scalable, real-time AD screening in diverse settings, including both high-resource and resource-constrained settings. Full article

Lipid vesicles and related nanocarriers often contain two compartments, such as the inner and outer leaflets of a bilayer membrane between which amphipathic molecules can migrate. We develop a stochastic model for describing the transfer kinetics of cargo between the two compartments in an ensemble of carriers, neglecting inter-carrier exchange to focus exclusively on intra-carrier redistribution. Starting from a set of rate equations, we examine the Gaussian regime in the limit of low cargo occupation where Gaussian and Poissonian statistics overlap. We derive a Fokker–Planck equation that we solve analytically for any initial cargo distribution among the carriers. Moments of the predicted distributions and examples, including a comparison between numerical solutions of the rate equations and analytic solutions of the Fokker–Planck equation, are presented and discussed, thereby establishing a theoretical foundation to study coupled intra- and inter-carrier transport processes in mobile nanocarrier systems. Full article

Grouting power serves as a critical parameter reflecting real-time energy input during grouting operations, and its accurate prediction is essential for intelligent control and engineering safety. Existing prediction methods often struggle to handle the strong nonlinearity, noise interference, adaptability to varying conditions in grouting power data. To address these challenges, an intelligent grouting system that integrates real-time data collection and core control modules has been developed. Subsequently, a grouting power prediction model is then proposed, which combines Complete Ensemble Empirical Mode Decomposition and Adaptive Noise (CEEMDAN) with a Convolutional Neural Net-work-Bidirectional Long Short-Term Memory Neural Network (CNN-BiLSTM) is proposed. The approach employs CEEMDAN to decompose the nonlinear and non-stationary power sequence into multiple intrinsic mode functions (IMFs). Each IMF is then separated into linear and nonlinear components using a moving average method. The linear components are predicted using an Autoregressive Integrated Moving Average (ARIMA) model, while the nonlinear components are predicted using a CNN-BiLSTM model. The final prediction is obtained by reconstructing the results from both components. Experimental comparisons under both normal and heaving grouting conditions demonstrate that the proposed model significantly outperforms LSTM, CNN-LSTM, and CNN-BiLSTM models. With 80% of the dataset used for training, the RMSE for normal conditions is reduced by 95.69%, 85.11%, and 80.55%, respectively, and for heaving conditions by 94.91%, 90.71%, and 84.62%, respectively. This research provides high-precision predictive support for grouting regulation under complex working conditions, offering substantial engineering application value. Full article

The objective of this study is to develop an artificial intelligence algorithm that can predict both the risk level and damage level of reinforced concrete structures through classification and proportioning. This algorithm identifies buildings that require preventive measures before an earthquake and buildings that require immediate repair or demolition after an earthquake. A key aspect of the approach is calculating each building’s risk level as the ratio of its risky story to the total number of stories. That calculation provides a normalized figure, enabling comparison between buildings of varying sizes and complexities in an equitable way. The dataset of this study includes 100 buildings affected by previous earthquakes in Türkiye and 782 buildings with detailed seismic analysis. Thirteen different building parameters, structural, seismic, and geometric, have been considered within the scope of this study. Rapid visual screening (RVS) methods were applied for structural integrity analysis, and machine learning models were used for improvement in accuracy and efficiency. In the comparison of the model sets, the approach achieved the highest accuracy of 77% with an ensemble of four models. The results demonstrate the value of blending AI with traditional methodologies for risk analysis. It shows a viable and scalable mechanism for prioritization of retrofit and inspections and helps engineers and policymakers enhance disaster preparedness. By identifying structures at high risk, this work contributes towards overall aims for earthquake resilience in buildings. This study introduces a Pearson-correlation-based feature analysis and a Random Oversampling strategy to enhance model balance. The ensemble model achieved 83% external accuracy and outperformed the traditional RVS method (68%), reducing computation time from minutes to seconds. Full article