Search Results (966)

Federated semi-supervised learning (Fed-SSL) has emerged as a powerful framework that leverages both labeled and unlabeled data distributed across clients. To reduce communication overhead, real-world deployments often adopt partial client participation, where only a subset of clients is selected in each round. However, under non-i.i.d. data distributions, the choice of client sampling strategy becomes critical, as it significantly affects training stability and final model performance. To address this challenge, we propose a novel federated averaging semi-supervised learning algorithm, called FedAvg-SSL, that considers two sampling approaches, uniform random sampling (standard Monte Carlo) and a structured lattice-based sampling, inspired by quasi-Monte Carlo (QMC) techniques, which ensures more balanced client participation through structured deterministic selection. On the client side, each selected participant alternates between updating the global model and refining the pseudo-label model using local data. We provide a rigorous convergence analysis, showing that FedAvg-SSL achieves a sublinear convergence rate with linear speedup. Extensive experiments not only validate our theoretical findings but also demonstrate the advantages of lattice-based sampling in federated learning, offering insights into the interplay among algorithm performance, client participation rates, local update steps, and sampling strategies. Full article

Precisely recognizing diseases in tomato leaves plays a crucial role in enhancing the health, productivity, and quality of tomato crops. However, disease identification methods that rely on single-mode information often face the problems of insufficient accuracy and weak generalization ability. Therefore, this paper proposes a tomato leaf disease recognition framework FCMNet based on multimodal fusion, which combines tomato leaf disease image and text description to enhance the ability to capture disease characteristics. In this paper, the Fourier-guided Attention Mechanism (FGAM) is designed, which systematically embeds the Fourier frequency-domain information into the spatial-channel attention structure for the first time, enhances the stability and noise resistance of feature expression through spectral transform, and realizes more accurate lesion location by means of multi-scale fusion of local and global features. In order to realize the deep semantic interaction between image and text modality, a Cross Vision–Language Alignment module (CVLA) is further proposed. This module generates visual representations compatible with Bert embeddings by utilizing block segmentation and feature mapping techniques. Additionally, it incorporates a probability-based weighting mechanism to achieve enhanced multimodal fusion, significantly strengthening the model’s comprehension of semantic relationships across different modalities. Furthermore, to enhance both training efficiency and parameter optimization capabilities of the model, we introduce a Multi-strategy Improved Coati Optimization Algorithm (MSCOA). This algorithm integrates Good Point Set initialization with a Golden Sine search strategy, thereby boosting global exploration, accelerating convergence, and effectively preventing entrapment in local optima. Consequently, it exhibits robust adaptability and stable performance within high-dimensional search spaces. The experimental results show that the FCMNet model has increased the accuracy and precision by 2.61% and 2.85%, respectively, compared with the baseline model on the self-built dataset of tomato leaf diseases, and the recall and F1 score have increased by 3.03% and 3.06%, respectively, which is significantly superior to the existing methods. This research provides a new solution for the identification of tomato leaf diseases and has broad potential for agricultural applications. Full article

The invasion of alien species has emerged as a global ecological challenge and invasive species can seriously threaten the habitats of native plants and intensify interspecific competition, ultimately exerting significant impacts on local ecosystems. Therefore, it is necessary to implement effective prevention and control strategies to reduce these impacts and maintain ecological stability. Against this backdrop, it is especially critical to analyze the influencing factors of invasive alien species and predict their future trends. Given China’s vast territory, complex natural geography, and diverse climatic conditions, the problem of invasive alien species in China is particularly severe, and scientific countermeasures are urgently required. Up to now, the number of invasive alien plant species in China has exceeded 520. Based on the number of invasive plant species in each province of China, this study analyzes the intrinsic connection between various influencing factors and invasive species, and through correlation analysis identifies the influencing factors, which are then used to analyze and predict the future invasion risks that each region may face. Full article

Fall detection is a critical task in healthcare monitoring systems, especially for elderly populations, for whom timely intervention can significantly reduce morbidity and mortality. This study proposes a privacy-preserving and scalable fall-detection framework that integrates federated learning (FL) with transfer learning (TL) to train deep learning models across decentralized data sources without compromising user privacy. The pipeline begins with data acquisition, in which annotated video-based fall-detection datasets formatted in YOLO are used to extract image crops of human subjects. These images are then preprocessed, resized, normalized, and relabeled into binary classes (fall vs. non-fall). A stratified 80/10/10 split ensures balanced training, validation, and testing. To simulate real-world federated environments, the training data is partitioned across multiple clients, each performing local training using pretrained CNN models including MobileNetV2, VGG16, EfficientNetB0, and ResNet50. Two FL topologies are implemented: a centralized server-coordinated scheme and a ring-based decentralized topology. During each round, only model weights are shared, and federated averaging (FedAvg) is applied for global aggregation. The models were trained using three random seeds to ensure result robustness and stability across varying data partitions. Among all configurations, decentralized MobileNetV2 achieved the best results, with a mean test accuracy of 0.9927, F1-score of 0.9917, and average training time of 111.17 s per round. These findings highlight the model’s strong generalization, low computational burden, and suitability for edge deployment. Future work will extend evaluation to external datasets and address issues such as client drift and adversarial robustness in federated environments. Full article

The structural health monitoring of historic buildings represents one of the most significant challenges in contemporary structural analysis, particularly for large-scale structures with accumulated damage. Obtaining reliable diagnostics is crucial yet complex due to the inherent uncertainties in both geometric definition and material properties of historic constructions, especially when structural stability may be compromised. This study presents a comprehensive structural assessment of the Bullring of the Real Maestranza de Caballería de Ronda (Spain), an emblematic 18th-century structure, through an innovative multi-technique approach aimed at evaluating its structural stability. The methodology integrates various non-destructive techniques: 3D laser scanning for precise geometric documentation, operational modal analysis (OMA) for global dynamic characterisation, experimental modal analysis (EMA) for local assessment of critical structural elements, and sonic tests (ST) to determine the elastic moduli of the principal materials that define the historic construction. The research particularly focuses on the inner ring of sandstone columns, identified as the most vulnerable structural component through initial dynamic testing. A detailed finite-element (FE) model was developed based on high-precision laser-scanning data and calibrated using experimental dynamic properties. The model’s reliability was validated through the correlation between numerical predictions and experimental observations, enabling a thorough stability analysis of the structure. Results reveal concerning stability issues in specific columns of the inner ring, identifying elements at significant risk of collapse. This finding demonstrates the effectiveness of the proposed methodology in detecting critical structural vulnerabilities in historic buildings, providing crucial information for preservation strategies. Full article

To address the challenges of image inpainting in scenarios with extensive or irregular missing regions—particularly detail oversmoothing, structural ambiguity, and textural incoherence—this paper proposes an Image Structure-Guided (ISG) framework that hierarchically integrates structural priors with semantic-aware texture synthesis. The proposed methodology advances a two-stage restoration paradigm: (1) Structural Prior Extraction, where adaptive edge detection algorithms identify residual contours in corrupted regions, and a transformer-enhanced network reconstructs globally consistent structural maps through contextual feature propagation; (2) Structure-Constrained Texture Synthesis, wherein a multi-scale generator with hybrid dilated convolutions and channel attention mechanisms iteratively refines high-fidelity textures under explicit structural guidance. The framework introduces three innovations: (1) a hierarchical feature fusion architecture that synergizes multi-scale receptive fields with spatial-channel attention to preserve long-range dependencies and local details simultaneously; (2) spectral-normalized Markovian discriminator with gradient-penalty regularization, enabling adversarial training stability while enforcing patch-level structural consistency; and (3) dual-branch loss formulation combining perceptual similarity metrics with edge-aware constraints to align synthesized content with both semantic coherence and geometric fidelity. Our experiments on the two benchmark datasets (Places2 and CelebA) have demonstrated that our framework achieves more unified textures and structures, bringing the restored images closer to their original semantic content. Full article

Sheath blight (ShB), caused by the necrotrophic fungus Rhizoctonia solani (R. solani), poses severe threats to global rice production. Developing a resistant variety with an ShB-resistance gene is one of most efficient and economical approaches to control the disease. Here, we identified a highly conserved chloroplast-localized stem-loop-binding protein encoding gene (OsCSP41b), which shows great potential in developing an ShB-resistant variety. OsCSP41b-knockout mutants exhibit chlorotic leaves and increased ShB susceptibility, whereas OsCSP41b-overexpressing lines (CSP41b-OE) display significantly enhanced resistance to R. solani, as well as to drought, and salinity stresses. Notably, CSP41b-OE lines present a completely comparable grain yield to the wild type (WT). Transcriptomic analyses reveal that chloroplast transcripts and photosynthesis-associated genes maintain observably elevated stability in CSP41b-OE plants versus WT plants following R. solani infection, which probably accounts for the enhanced ShB resistance of CSP41b-OE. Our findings nominate the OsCSP41b gene as a promising molecular target for developing a rice variety with stronger resistance to both R. solani and multi-abiotic stresses. Full article

A concentric double-flange butterfly valve (DN-500, PN-10) was analyzed to examine its dynamic behavior and internal fluid flow across multiple opening angles. Finite Element Analysis (FEA) was employed to determine natural frequencies, mode shapes, and effective mass participation factors (EMPFs) for valve positions at 30°, 60°, and 90°. The valve geometry was discretized using a curvature-based mesh with linear elastic isotropic properties for 1023 carbon steel. Lower-order vibration modes produced global deformations primarily along the valve disk, while higher-order modes showed localized displacement near the shaft–bearing interface, indicating coupled torsional and translational dynamics. The highest EMPF in the X-direction occurred at 1153.1 Hz with 0.2631 kg, while the Y-direction showed moderate contributions peaking at 0.1239 kg at 392.06 Hz. The Z-direction demonstrated lower influence, with a maximum EMPF of 0.1218 kg. Modes 3 and 4 were critical for potential resonance zones due to significant mass contributions and directional sensitivity. Computational Fluid Dynamics (CFD) simulation analyzed flow behavior, pressure drops, and turbulence under varying valve openings. At a lower opening angle, significant flow separation, recirculation zones, and high turbulence were observed. At 90°, the flow became more streamlined, resulting in a reduction in pressure losses and stabilizing velocity profiles. Full article

In the drilling process of ultra-deep wells with large-diameter boreholes, the transport and deposition behavior of cuttings plays a critical role in maintaining wellbore cleanliness and ensuring operational safety. Due to the geometry of enlarged boreholes and their complex annular flow characteristics, conventional single-parameter control methods often fail to achieve effective cuttings transport. This study aims to identify the dominant influencing factors and optimize key parameters by focusing on the cuttings volume fraction as a primary evaluation metric. A numerical simulation approach is employed to systematically investigate the influence of stabilizer geometry and hydraulic parameters. Five variables—drilling fluid velocity, drill pipe rotational speed, number of stabilizers, flow area, and helical angle—are selected for analysis. An initial one-factor sensitivity analysis is conducted to evaluate local impacts and to establish relative sensitivity indices, thereby identifying key variables. A variance-based global sensitivity analysis is further applied to quantify first-order effects, full-order effects, and interaction contributions, revealing nonlinear coupling and synergistic mechanisms. The results indicate that drilling fluid velocity and rotation speed exhibit the most significant first-order influences, while stabilizer-related parameters show strong interaction effects that are often underestimated by traditional methods. Based on these findings, an optimized cuttings transport scheme for large-diameter boreholes is proposed. Additionally, a multi-parameter response model for the cuttings volume fraction is developed using sensitivity-weighted analysis, offering theoretical support and methodological reference for enhancing cuttings transport performance and structural design in large-diameter borehole drilling operations. Full article

National parks represent a global initiative for biodiversity conservation and environmentally sustainable societal development, with China having launched its own national park program. The establishment and operation of these parks significantly impact local residents’ livelihoods. Based on DFID’s Sustainable Livelihoods Framework, an assessment tool introduced by the UK Department for International Development (DFID) for evaluating the livelihood standards of residents, this study constructs a structural equation modeling (SEM) framework to analyze how national parks affect residents’ livelihoods, discussing livelihood risk management and feasible capacity-building interventions. Focusing on the Northeast Tiger and Leopard National Park as a case study, the research reveals that indirect wildlife-inflicted damage poses more pronounced negative impacts on local communities than park establishment policies. Both regulatory land-use restrictions and wildlife conflicts disrupt land-based livelihood activities, ultimately affecting residents’ livelihood stability. Mitigation requires comprehensive measures, including retaining essential farmland; providing vocational skill training; offering specialized loans; diversifying employment channels; and improving compensation mechanisms to safeguard residents’ livelihood security. Full article

This paper proposes a Distributed Adaptive Regularization Algorithm (DARN) for solving composite non-convex and non-smooth optimization problems in multi-agent systems. The algorithm employs a three-phase iterative framework to achieve efficient collaborative optimization: (1) a local regularized optimization step, which utilizes proximal mappings to enforce strong convexity of weakly convex objectives and ensure subproblem well-posedness; (2) a consensus update based on doubly stochastic matrices, guaranteeing asymptotic convergence of agent states to a global consensus point; and (3) an innovative adaptive regularization mechanism that dynamically adjusts regularization strength using local function value variations to balance stability and convergence speed. Theoretical analysis demonstrates that the algorithm maintains strict monotonic descent under non-convex and non-smooth conditions by constructing a mixed time-scale Lyapunov function, achieving a sublinear convergence rate. Notably, we prove that the projection-based update rule for regularization parameters preserves lower-bound constraints, while spectral decay properties of consensus errors and perturbations from local updates are globally governed by the Lyapunov function. Numerical experiments validate the algorithm’s superiority in sparse principal component analysis and robust matrix completion tasks, showing a 6.6% improvement in convergence speed and a 51.7% reduction in consensus error compared to fixed-regularization methods. This work provides theoretical guarantees and an efficient framework for distributed non-convex optimization in heterogeneous networks. Full article

As a core component of automated logistics systems, delivery robots hold significant application value in the field of unmanned delivery. This research addresses the robot path planning problem, aiming to enhance delivery efficiency and reduce operational costs through systematic improvements to the RIME optimization algorithm. Through in-depth analysis, we identified several major drawbacks in the standard RIME algorithm for path planning: insufficient global exploration capability in the initial stages, a lack of diversity in the hard RIME search mechanism, and oscillatory phenomena in soft RIME step size adjustment. These issues often lead to undesirable phenomena in path planning, such as local optima traps, path redundancy, or unsmooth trajectories. To address these limitations, this study proposes the Multi-Strategy Controlled Rime Algorithm (MSRIME), whose innovation primarily manifests in three aspects: first, it constructs a multi-strategy collaborative optimization framework, utilizing an infinite folding Fuch chaotic map for intelligent population initialization to significantly enhance the diversity of solutions; second, it designs a cooperative mechanism between a controlled elite strategy and an adaptive search strategy that, through a dynamic control factor, autonomously adjusts the strategy activation probability and adaptation rate, expanding the search space while ensuring algorithmic convergence efficiency; and finally, it introduces a cosine annealing strategy to improve the step size adjustment mechanism, reducing parameter sensitivity and effectively preventing path distortions caused by abrupt step size changes. During the algorithm validation phase, comparative tests were conducted between two groups of algorithms, demonstrating their significant advantages in optimization capability, convergence speed, and stability. Further experimental analysis confirmed that the algorithm’s multi-strategy framework effectively suppresses the impact of coordinate and dimensional differences on path quality during iteration, making it more suitable for delivery robot path planning scenarios. Ultimately, path planning experimental results across various Building Coverage Rate (BCR) maps and diverse application scenarios show that MSRIME exhibits superior performance in key indicators such as path length, running time, and smoothness, providing novel technical insights and practical solutions for the interdisciplinary research between intelligent logistics and computer science. Full article

Skin cancer is one of the most prevalent forms of cancer worldwide, and early and accurate diagnosis critically impacts patient outcomes. Given the sensitive nature of medical data and its fragmented distribution across institutions (data silos), privacy-preserving collaborative learning is essential to enable knowledge-sharing without compromising patient confidentiality. While federated learning (FL) offers a promising solution, existing methods struggle with heterogeneous and missing modalities across institutions, which reduce the diagnostic accuracy. To address these challenges, we propose an effective and flexible Personalized Multimodal Federated Learning framework (PMM-FL), which enables efficient cross-client knowledge transfer while maintaining personalized performance under heterogeneous and incomplete modality conditions. Our study contains three key contributions: (1) A hierarchical aggregation strategy that decouples multi-module aggregation from local deployment via global modular-separated aggregation and local client fine-tuning. Unlike conventional FL (which synchronizes all parameters in each round), our method adopts a frequency-adaptive synchronization mechanism, updating parameters based on their stability and functional roles. (2) A multimodal fusion approach based on multitask learning, integrating learnable modality imputation and attention-based feature fusion to handle missing modalities. (3) A custom dataset combining multi-year International Skin Imaging Collaboration(ISIC) challenge data (2018–2024) to ensure comprehensive coverage of diverse skin cancer types. We evaluate PMM-FL through diverse experiment settings, demonstrating its effectiveness in heterogeneous and incomplete modality federated learning settings, achieving 92.32% diagnostic accuracy with only a 2% drop in accuracy under 30% modality missingness, with a 32.9% communication overhead decline compared with baseline FL methods. Full article

Housing prices have recently risen sharply in many countries, primarily linked to the global credit cycle. Although various factors play a role, the ability of developing countries to navigate this cycle and maintain autonomous monetary policies is crucial. This paper introduces a dynamic macroeconomic model featuring a housing production sector within an imperfect banking framework. It captures key housing and economic dynamics in advanced and emerging economies. The analysis shows domestic liquidity policies, such as bank capital requirements, reserve ratios, and currency devaluation, can stabilize investment and production. However, their effectiveness depends on foreign interest rates and liquidity. Stabilizing housing prices and risk-free bonds is more effective in high-interest environments, while foreign liquidity shocks have asymmetric impacts. They can boost or lower the effectiveness of domestic policy, depending on the country’s level of financial development. These findings have several policy implications. For example, foreign capital controls would be adequate in the short term but not in the long term. Instead, governments would try to promote the development of local financial markets. Controlling debt should be a target for macroprudential policy as well as promoting saving instruments other than real estate, especially during low interest rates. Full article

Background/Objectives: Understanding the molecular interactions between small bioactive compounds and serum albumins is essential for drug development and pharmacokinetics. Noraucuparin, a biphenyl-type phytoalexin with promising pharmacological properties, has shown a strong binding affinity to bovine serum albumin (BSA), a model protein for drug transport. This study aims to elucidate the structural and energetic characteristics of the noraucuparin–BSA complex under physiological and slightly elevated temperatures. Methods: Microsecond-scale molecular dynamics (MD) simulations and Molecular Mechanics Generalized Born Surface Area (MMGBSA)-binding-free energy calculations were performed to investigate the interaction between noraucuparin and BSA at 298 K and 310 K. Conformational flexibility and per-residue energy decomposition analyses were conducted, along with interaction network mapping to assess ligand-induced rearrangements. Results: Noraucuparin preferentially binds to site II of BSA, near the ibuprofen-binding pocket, with stabilization driven by hydrogen bonding and hydrophobic interactions. Binding at 298 K notably increased the structural mobility of BSA, affecting its global conformational dynamics. Key residues, such as Trp213, Arg217, and Leu237, contributed significantly to complex stability, and the ligand induced localized rearrangements in the protein’s intramolecular interaction network. Conclusions: These findings offer insights into the dynamic behavior of the noraucuparin–BSA complex and enhance the understanding of serum albumin–ligand interactions, with potential implications for drug delivery systems. Full article