ABSTRACT Statistical process control (SPC) methods are commonly employed in various fields to detect distributional changes in sequential processes. Traditional SPC charts are typically developed under the assumption that in‐control (IC) process observations are independent and normally distributed with identical parameters. However, when these assumptions are violated, recent research has shown that conventional control charts may become unreliable. To address these limitations, various alternative and flexible control charts have been developed to accommodate autocorrelated observations and nonparametric process distributions. Although existing methods can be reliable and effective when their assumptions hold, they still have some limitations. For instance, methods handling autocorrelated data often rely on parametric time series models or assume equally spaced observations, whereas nonparametric control charts that rely on data ranking or categorization typically suffer from information loss. Furthermore, the optimal performance of many control charts in detecting specific shifts often relies on the accurate specification of their parameters in advance. In this paper, we introduce a novel framework for Phase II online monitoring of univariate processes with irregularly spaced observation times and serial correlation, and the IC distribution cannot be adequately modeled by a parametric form. The method first estimates the IC covariance function for irregularly spaced time series using a local linear kernel smoothing procedure, then sequentially decorrelates the process observations. Next, the decorrelated observations are transformed based on their estimated IC distribution such that the transformed data are approximately standard normal. Finally, an adaptive CUSUM chart is employed to monitor the transformed data. Simulation results indicate that the proposed approach is effective across a variety of scenarios.

Tuberculosis represents a major infectious pathology attributed to the Mycobacterium tuberculosis bacterium (MTB), and delayed detection can lead to serious consequences or death. Existing methods suffer from limited accuracy, poor bacilli segmentation and inadequate feature extraction. To address these issues, a Siberian Tiger Parrot Optimization-powered Deep High-Order Attention Model (STPO_DHA-Net) is proposed for tuberculosis severity-levels from sputum images. STPO is the hybridization of Siberian Tiger Optimization (STO) and Parrot Optimizer (PO), used to tune the parameters of DHA-Net. Initially, a high-boost filter enhances fine structural information. Then, the Shape-aware Loss-based Additive Manufacturing SegNet (AM-SegNet) precisely segments the bacilli. Then, discriminative features are derived using Local Adaptive Regression Kernels (LARK), complemented by Haralick texture descriptors, including entropy, uniformity, contrast, homogeneity and correlation. DHA-Net identifies severity level by focusing on subtle microbial-level patterns. STPO_DHA-Net attained 91.88% accuracy, 92.33% precision, 92.64% True Negative Rate (TNR), 91.65% F1-score and 90.67% True Positive Rate (TPR).

Classic machine learning and regime identification methods applied to financial time series lack theoretical guarantees and exhibit systematic failure modes: heavy-tails invalidate moment-based geometry, rendering distances and centroids dominated by extremes or unstable; jumps violate smoothness, destabilizing local regressions, kernel methods, and gradient-based learning; and non-stationarity disrupts neighborhood relations, so distances in classical feature spaces no longer reflect meaningful proximity. To address these challenges, we propose a topology-based machine-learning framework grounded on probabilistic reconstruction of state-space geometry, which replaces moment- and smoothness-dependent representations with deformation-stable summaries of state-space geometry, preserving neighborhoods, adjacency, and topology. The finite-sample validity of homeomorphic state-space reconstruction, required for topology-based machine learning, is assessed through numerical studies on synthetic data with heavy tails, jumps, and known ground-truth regimes. Further diagnostics of local invertibility and bounded geometric distortion quantify when embedding windows are consistent with local diffeomorphic behavior, enabling metric-sensitive, geometry-aware learning. Clustering of Hilbert-space summaries accurately recovers underlying market tail-risk regimes with robust results across selected filtrations. Temporal, feature-space, and cluster-label null tests confirm that topology-based clustering captures genuine topological structure rather than noise or artifacts, and encodes temporal dependencies at local, mesoscopic, and network levels associated with market regimes.

In modern survey sampling, particularly when using stratified random sampling (StRS), the existence of outliers and model mis-specifications is a daunting challenge to the conventional parametric and nonparametric methods of estimating parameters. This research presents a new type of predictive estimator that is synergistic to both robust regression and nonparametric local polynomial kernel regression. It aims to offer more resistant and efficient estimators of the average parameter in the areas where supplementary information is known, but irregularity in the data is usual. The proposed estimators use dual calibration methods based on both auxiliary variable means and coefficients of variation, which improves efficiency. This framework enhances predictive performance by integrating the adaptability of kernel-based smoothing with the outlier resistance of robust regression. The accuracy of the suggested estimators is measured by using large scales of simulation experiments on artificial populations with structural heterogeneity and outlier contamination. An empirical comparison, based on percentage relative efficiency (PRE), indicates that the new estimators are superior to classical methods based on the use of a kernel regression in most bandwidth selection strategies. In addition to bringing methodological innovation as it connects distribution theory, regression models, and robust estimation strategies, this work also offers the usefulness of survey practitioners who work with complicated and imperfect real-life data of fisheries and radiations.

Diabetes mellitus poses a growing global health concern, particularly in low- and middle-income countries where early detection remains limited, demanding classification models that balance accuracy, interpretability, and adaptability to heterogeneous clinical data. This study proposes and evaluates the Adaptive Kernel Probability Model (AKPM), a novel nonparametric probabilistic classifier designed to enhance diabetes prediction by performing localized kernel density estimation with adaptive bandwidth selection via k-nearest neighbors. Implemented and tested on the Pima Indians Diabetes Dataset, AKPM outperformed conventional classifiers—Naïve Bayes and Gaussian Mixture Models (GMM)—across all evaluation metrics, achieving 87.5% accuracy, 83.3% precision, 76.9% recall, and an F1-score of 80.0% for the diabetic class, alongside 89.3% precision and 92.6% recall for the normal class. These results surpassed GMM (83.0% accuracy, 71.6% F1-score) and Naïve Bayes (80.0% accuracy, 66.6% F1-score), confirming AKPM’s superior capability to detect diabetic cases while minimizing false negatives. Offering transparent posterior inference and a modular design, AKPM emerges as a reliable and interpretable solution for clinical decision support systems and real-world healthcare applications.

During the fruit harvesting and thinning processes, natural wind forces and robotic pull‐based harvesting or thinning methods often induce fruit oscillation, which subsequently complicates robot localization and reduces the success rate of the operation. To address this issue, this article proposes a solution that combines the CHD‐YOLOv8n detection model with an optimized DeepSORT tracking algorithm. CHD‐YOLOv8n is an improved version of the YOLOv8n model. First, in response to the computational limitations of mobile devices, we introduce the CAA‐HSFPN (Channel Attention Adaptive Hybrid Spatial Feature Pyramid Network) to replace the original head network in YOLOv8n. This modification dynamically optimizes feature weight distribution, reducing the number of parameters and improving feature transmission efficiency. Second, to address the degradation in recognition caused by leaf occlusion, the C2f‐DAttention (C2f Dual Attention) module is proposed, which utilizes both spatial and channel attention mechanisms to enhance feature extraction in the fruit region. Finally, to further improve fruit detection accuracy, the SPPF‐LSKA (Spatial Pyramid Pooling with Local Spatial Kernel Attention) module is incorporated, enabling the model to effectively fuse features at multiple scales. To improve the tracking performance of DeepSORT in orchard environments, the original detector with CHD‐YOLOv8n is replaced, and the ResNest50 network is introduced to enhance the discriminative power of appearance features. Meanwhile, CIoU is used to replace traditional IoU matching, optimizing the dynamic association of fruits. To reduce identity loss in oscillating fruit scenes, an adaptive noise‐scale Kalman filter is designed. The experimental results show that the CHD‐YOLOv8n model achieves mAP@0.5 of 95.23% and 96.18% for detecting young and mature peaches, respectively, with both precision and recall exceeding 91%. When combined with theoptimized DeepSORT algorithm, the tracking accuracy improves by 13.2–16.5% compared to traditional SORT, while the number of ID switches is reduced by 50–59.46%. These technical innovations provide an efficient and stable solution for intelligent harvesting and thinning robots.

Abstract. Data assimilation involves sequential inference in geophysical systems with nonlinear dynamics and observational operators. Non-parametric filters are a promising approach for data assimilation because they are able to represent non-Gaussian densities. The mapping particle filter is an iterative ensemble method that incorporates the Stein Variational Gradient Descent (SVGD) to produce a particle flow transforming state vectors from prior to posterior densities. At every pseudo-time step, the Kullback-Leibler divergence between the intermediate density and the target posterior is evaluated and minimized. However, for applications in geophysical systems, challenges persist in high dimensions, where sample covariance underestimation leads to filter divergence. This work proposes two localization methods, one in which a local kernel function is defined and the particle flow is global. The second method, given a localization radius, physically partitions the state vector and performs local mappings at each grid point. The performance of the proposed Local Mapping Particle Filters (LMPFs) is assessed in synthetic experiments. Observations are produced with a two-scale Lorenz system, while a one-scale Lorenz model is used as surrogate, introducing model error in the inference. The methods are evaluated with both full and partial observations, as well as with different linear and non-linear observational operators. The LMPFs with Gaussian mixtures in the prior density perform similarly to Gaussian filters such as the Ensemble Transform Kalman Filter (ETKF) and the Local Ensemble Transform Kalman Filter (LETKF) in most cases, and in some scenarios, they provide competitive performance in terms of analysis accuracy.

Abstract Assessment of a continuous spatial variable based on a set of m observations is usually performed in a Gaussian random field framework. The optimal predictor under this model can be presented either as a linear kriging predictor or as a dual kriging predictor. The spatial variable predictor is usually stored in a kriging grid representation of size n . Alternatively, one may define a kernel function representation based on the dual kriging formulation. The latter can be efficiently reduced to the former, but not vice versa. To provide a prediction at an arbitrary location, a piecewise planar interpolation in the actual grid unit is typically required. For the functional representation, the functional value in the actual location must be calculated. The computational challenge of both representations is primarily related to the inversion of the observation covariance $$ ( m \times m ) $$ ( m × m ) -matrix. In large spatial studies with huge sets of observations, and thus huge m , this inversion may not be computationally feasible. Localized kriging predictors are then frequently used to generate the grid representation of the spatial variable. This approach has computational demands proportional to the grid size n . We present a localized kernel predictor to provide a functional representation of the spatial variable. The specification of this localized kernel predictor constitutes the major contribution of this paper. This predictor has computational demands proportional to the number of observations m . This is particularly beneficial in three-dimensional models and spatiotemporal studies where one typically has $$ n \gg m $$ n ≫ m . The characteristics of the kernel predictor are demonstrated in an example. A study on real observations indicates that the localized kernel function representation has substantial computational advantages over the localized kriging grid representation. Even generating the grid representation from a kernel function representation appears more computationally efficient than generating it directly using a localized kriging predictor.

Decoding speech intentions from electroencephalogram (EEG) data is the primary task in speech brain-computer interface (BCI) systems, which remains challenging due to the unclear discriminative task-aware features, and underlying nonlinear properties besides the well-known low signal-to-noise ratio of EEG data. Existing approaches typically rely either on single-domain features or performing feature learning by deep neural networks; therefore, they either fail to capture comprehensive signal patterns, or typically require large-sized EEG data to fit the parameter spaces and often have limited interpretability. To address these limitations, we propose a Multi-view Manifold-Adaptive Kernel Regression (MMKR) model for speech recognition from EEG signals in this paper. By treating temporal, spectral, and statistical EEG representations as complementary feature views, view-specific manifold-adaptive kernels are constructed in MMKR to incorporate local graph structure into kernel similarity; besides, a data-driven adaptive view weighting mechanism is used to characterize their contributions. We evaluate MMKR on both overt and imagined speech EEG datasets and the results demonstrate that MMKR achieves superior classification accuracy and robustness compared to some representative single-view, multi-view, and kernel-based baselines. Moreover, analysis on the local manifold-modulated kernel matrix and the learned view contributions are provided.

Sleep instability is a typical characteristic of insomnia, manifested as the inability of the brain to maintain a stable state, but its precise quantification is still challenging. We assume that sleep instability fundamentally reflects an increase in unpredictability in the evolution of brain network dynamics. To verify this, an interval prediction framework combining the temporal mobile network (TMN) and local adaptive kernel density estimation (LAKDE) is proposed to characterize the sleep instability. Specifically, TMN predicts future network states, while LAKDE module quantifies the uncertainty of these predictions by generating prediction intervals (PIs). Experiments on SIESTA and Sleep-EDF databases have shown that this method can construct well calibrated PIs. The key finding is that the PI normalized average width (PINAW) of subjects with sleep disorders is significantly higher than that of the healthy control group, validating that wider PIs are a mechanistic biomarker of sleep instability. In addition, this study further revealed a significant correlation between PINAW and traditional indicators such as number of sleep stage transitions, indicating that dynamic instability based on prediction uncertainty shares a common physiological basis with sleep fragmentation phenomena, establishing interval prediction as a paradigm for quantifying sleep instability.

IntroductionCerebral microbleeds (CMBs) are small hemorrhagic lesions visible as hypointense foci on susceptibility-sensitive MRI and are established biomarkers of stroke risk and amyloid-related imaging abnormalities (ARIA-H) in patients receiving anti-amyloid therapy. However, automated detection remains challenging because true CMBs closely resemble veins, calcifications, and susceptibility artifacts. This visual ambiguity results in a persistent precision–recall trade-off, where models optimized for high sensitivity tend to generate excessive false positives, while precision-focused models risk missing clinically relevant lesions. To address this limitation, we propose an attention-enhanced segmentation framework designed to suppress confounding activations while preserving lesion sensitivity.MethodsWe developed RLK-UNet with Convolutional Block Attention Modules (CBAM), a single-stage encoder–decoder architecture that redefines skip connections as context-filtered pathways. The encoder incorporates large 13×13 residual local kernel (RLK) convolutions to capture broad contextual information for distinguishing spherical microbleeds from elongated vascular structures. CBAM modules are embedded in all skip connections to selectively enhance lesion-relevant features and suppress irrelevant background responses before feature fusion. The model was trained and evaluated on a multi-site dataset of 506 T2*-GRE and SWI scans, with lesion-level detection assessed using precision, recall, F1-score, and average false positives per scan. Subject-level burden estimation was further evaluated using ARIA-H severity intervals.ResultsThe proposed model achieved state-of-the-art lesion-level performance, with a precision of 0.891, recall of 0.887, F1-score of 0.887, and a markedly reduced false positive rate of 0.83 per subject. Five-fold cross-validation demonstrated stable performance with minimal variance across splits. In lesions ≤3 mm, the model maintained strong detection performance (F1-score 0.869) while effectively controlling false positives. Cross-modality evaluation between T2*-GRE and SWI confirmed robust generalization. Ablation studies verified that CBAM significantly improved precision while preserving sensitivity, and Grad-CAM visualizations demonstrated more spatially focused and clinically interpretable attention patterns. Subject-level CMB counts strongly correlated with ground truth (Spearman ρ = 0.93), and severity classification aligned with ARIA-H intervals.ConclusionRLK-UNet with CBAM provides a robust and interpretable solution for automated CMB detection by directly addressing false-positive propagation through attention-guided skip connections. The framework achieves balanced precision and sensitivity within a single-stage architecture and demonstrates reliable subject-level burden estimation aligned with clinically meaningful ARIA-H categories. These findings support its potential application in vascular risk stratification and treatment monitoring in patients undergoing anti-amyloid therapy.

Pipeline systems play a crucial role in transporting fluids and gases across industrial infrastructures; however, detecting and classifying defects in these pipelines is essential to ensure safety, reliability, and uninterrupted operations. In this study, we employ the latest YOLOv11 deep learning model for automated detection of six common types of pipeline defects: Deformation, Obstacle, Rupture, Disconnect, Misalignment, and Deposition. A custom dataset of 1,500 images was prepared, where 900 images (60%) were used for training comprising 150 images per class and 600 images (40%) were reserved for validation, with 100 images per class. The YOLOv11 model demonstrated strong detection capability, achieving an overall accuracy of 91.77%. To further enhance performance, we integrated and compared three attention mechanisms: Self-Attention (SA), Local Kernel Attention (LKA), and Convolutional Block Attention Module (CBAM). The results showed that YOLOv11 + SA achieved the highest accuracy of 98.95%, followed by YOLOv11 + LKA with 98.54%, while YOLOv11 + CBAM reached 89.60%. These findings highlight that integrating attention mechanisms can significantly improve the defect detection accuracy of YOLOv11. Future work will focus on extending the dataset.

Abstract Motivated by a number of applications in signal processing, we study the following question. Given samples of a multidimensional signal of the form $$f(\varvec{\ell })=\sum _{k=1}^K a_k\exp (-i\langle \varvec{\ell }, \textbf{w}_k\rangle ), \quad \textbf{w}_1,\cdots ,\textbf{w}_k\in \mathbb {R}^q, \ \varvec{\ell }\in \mathbb {Z}^q, \ |\varvec{\ell }| <n,$$ determine the values of the number K of components, and the parameters $$a_k$$ and $$\textbf{w}_k$$ ’s. We note that the the number of samples of f in the above equation is $$(2n-1)^q$$ . We develop an algorithm to recuperate these quantities accurately using only a subsample of size $$\mathcal {O}(qn)$$ of this data. For this purpose, we use a novel localized kernel method to identify the parameters, including the number K of signals. Our method is easy to implement, and is shown to be stable under a very low SNR range. We demonstrate the effectiveness of our resulting algorithm using 2 and 3 dimensional examples from the literature, and show substantial improvements over state-of-the-art techniques including Prony based, MUSIC and ESPRIT approaches.

We present new convergence analyses for parallel subspace correction methods for unconstrained semicoercive and nearly semicoercive convex optimization problems, generalizing the theory of singular and nearly singular linear problems to a class of nonlinear problems. Our results demonstrate that the elegant theoretical framework developed for singular and nearly singular linear problems can be extended to unconstrained semicoercive and nearly semicoercive convex optimization problems. For semicoercive problems, we show that the convergence rate can be estimated in terms of a seminorm stable decomposition over the subspaces and the kernel of the problem, aligning with the theory for singular linear problems. For nearly semicoercive problems, we establish a parameter-independent convergence rate, assuming the kernel of the semicoercive part can be decomposed into a sum of local kernels, which aligns with the theory for nearly singular problems. To demonstrate the applicability of our results, we provide convergence analyses of two-level additive Schwarz methods for solving certain nonlinear partial differential equations with Neumann boundary conditions, within the proposed abstract framework.

This article presents an efficient parallel algorithm for LU factorization of large dense matrices based on a two-dimensional block-cyclic data distribution designed for multi-GPU computing environments. The proposed methodology addresses key challenges of large-scale linear algebra computations, including load balancing, minimizing communication overhead, and maximizing computational locality. By distributing matrix blocks cyclically across a GPU process grid, the algorithm ensures uniform workload distribution even for very large problem sizes, thereby avoiding idle time and reducing synchronization delays. The implementation leverages state-of-the-art GPU computing technologies, including CUDA streams, cuBLAS and cuSolver libraries for local numerical kernels, and NCCL for high-performance collective communications using NVLink interconnects. A look-ahead scheduling strategy and overlapping of communication with computation further increase the degree of parallelism, enabling sustained high utilization of GPU resources throughout the factorization pipeline. A detailed theoretical performance model is developed to analyze speedup, scalability, communication costs, and the impact of block size on total execution time. Numerical experiments conducted on an 8-GPU node with NVIDIA RTX 2080 Ti GPUs demonstrate excellent strong scaling, achieving up to 95 % parallel efficiency for matrices of size up to N = 20,000. The results confirm that the proposed multi-GPU LU factorization approach closely approaches the theoretical performance limits and significantly outperforms traditional CPU-based and hybrid CPU–GPU schemes. The method is highly suitable for large-scale scientific and engineering applications requiring fast and robust solution of linear systems, including computational fluid dynamics, structural mechanics, numerical simulation of physical processes, and machine learning workloads. Future research directions include extensions to sparse matrices with adaptive load balancing, mixed-precision acceleration with error correction, and generalization to other matrix factorizations such as Cholesky, QR, and LDL?. Keywords: LU factorization, multi-GPU systems, block-cyclic data distribution, parallel computing, CUDA, cuBLAS, cuSolver, NCCL, high-performance computing, scalability.

Abstract For recurrent event data analysis, covariates may have different influence on recurrence. Some of them have a linear effect on the occurrence of events, while others may exhibit additional complex nonlinear effects. Based on this fact, a semiparametric partial nonparametric rate model was proposed. This model can simultaneously accommodate the linear and nonlinear effects of covariates, thereby better adapting to the complexity of the data. For non-linear covariate effect functions, local polynomial kernel estimation was used. Then, profile likelihood was utilized to get the estimate of linear parametric. The large sample behaviour of the estimates was proposed and proved. A lot of numerical simulations were performed to evaluate the manifestations of the estimation. The results demonstrated the estimation yielded satisfactory performance.

Kernel Search Optimization (KSO) is characterized by insufficient accuracy in local search, which makes it difficult to achieve local optimization. Therefore, this paper proposes a Large Local Search Kernel Search Optimization (LLSKSO) to enhance the local optimization ability. LLSKSO achieves the performance improvement by introducing several strategies. First, the initial population is homogenized using the good point set mechanism. Then, the little dung beetle search mechanism of the Dung Beetle Optimizer (DBO) is introduced to enhance the local search capability of the KSO. Finally, the Cauchy-Gaussian mutation strategy is utilized to prevent the algorithm from falling into local traps. These three steps enable LLSKSO to achieve a dynamic balance between local and global search. In addition, to verify the performance and robustness of LLSKSO, comparison experiments between LLSKSO and 10 well-known algorithms are conducted on 50 benchmark test functions. From the statistical results of mean, best and variance of different algorithms, the LLSKSO algorithm outperforms the other algorithms. Finally, LLSKSO is applied to the engineering problem of carbon fiber drafting ratio optimization. Moreover, the experimental results obtained by LLSKSO yielded smaller line densities and greater strengths compared to other algorithms. LLSKSO achieves theoretical optima in 16 out of 20 high-dimensional benchmark functions, with an average CPU runtime reduced by 30% compared to baseline methods. Therefore, it can be shown that LLSKSO can be used as an effective optimization algorithm and engineering assistance.

Graph-based multi-view clustering has gained significant attention in recent years due to its superior ability to reveal clustering structures. However, existing methods often incur high computational costs when capturing local information and overlook the higher-order correlations between multiple views. To address these issues, we propose Tensorized Consensus Graph Learning for Incomplete Multi-View Clustering with Confidence Integration (TCGL). This approach constructs adjacency and local heat kernel graphs by filtering missing samples to better capture local structures while leveraging a t-SVD-based weighted tensor nuclear norm sparsification method to reduce noise. Additionally, we introduce a matrix energy-based adjacency graph normalization strategy that utilizes common nearest neighbors to generate probability matrices, enhancing noise resistance and improving structural exploration. Experimental results demonstrate that TCGL effectively handles incomplete data and significantly outperforms state-of-the-art approaches across multiple datasets.

The partial linear model is a very important class of semiparametric model in applied quantitative sciences. This article considers the estimation of a partial linear model with a discontinuous unknown non parametric function. We embed the jump-preserving techniques in the profiled local linear kernel smoothing method, then propose an adaptive jump-preserving profiled local linear estimation procedure to estimate the parametric coefficients and non parametric function. This method can automatically accommodate possible jumps of the non parametric function without knowing the number and locations of jump points. The resulting estimators can preserve the jumps well and also give smooth estimates of the continuity part. The asymptotical properties of the resulting estimators are demonstrated under some mild conditions. Several numerical simulations are conducted to evaluate the finite sample performance of the proposed methodologies.

This study aims to analyze the glycemic index (GI) value of boiled sweet corn (Zea mays saccharata Sturt.) and local corn (Zea mays) in male white rats (Rattus norvegicus). The method used is measuring blood glucose levels at 0, 30, 60, 90, and 120 minutes after sample administration, then calculating the area under the curve (AUC) to obtain the GI value. The results showed that boiled sweet corn kernels had a GI value of 74.34% and boiled local corn kernels 75.28%, both of which are classified as high glycemic index. This high GI value is influenced by starch content, dietary fiber content, and processing methods. Based on these results, it is recommended that people with diabetes mellitus limit their consumption of these two types of boiled corn