Search Results (560)

Reinforcement learning for robotic manipulation is often limited by poor sample efficiency and unstable training dynamics, challenges that are further amplified in federated settings due to data privacy constraints and task heterogeneity. To address these issues, we propose F-DRL,a federated dynamics-aware representation learning framework that enables multiple robotic tasks to collaboratively learn structured latent representations without sharing raw trajectories or policy parameters. The framework combines robotics priors with an action-conditioned latent dynamics model to learn low-dimensional state and state–action embeddings that explicitly capture task-relevant geometric and transition structure. Representation learning is performed locally at each client, while a central server aggregates encoder parameters using a similarity-weighted scheme based on second-order latent geometry. The learned representations are then used as frozen auxiliary inputs for downstream model-free reinforcement learning. We evaluate F-DRL on seven heterogeneous robotic manipulation tasks from the MetaWorld benchmark. While achieving performance comparable to centralized training and standard federated baseline, F-DRL substantially improves training stability relative to FedAvg on heterogeneous manipulation tasks with partially shared dynamics (e.g., Drawer-Open and Window-Open), reducing the mean across-seed standard deviation and the AUC of this deviation by over 60%. The method remains neutral on simple tasks and performs less consistently on contact-rich manipulation tasks with task-specific dynamics, indicating both the benefits and the practical limits of representation-level knowledge sharing in federated robotic learning. Full article

Decarbonization of the concrete industry has arisen as one of the main priorities for the construction sector in order to mitigate the negative climate impact associated with construction. The carbon emissions of concrete mainly originate from the production of cement, and it is essential to find supplementary cementitious materials (SCMs) to achieve eco-friendly construction materials. The use of tailings as SCMs could reduce the carbon footprint of concrete, as well as improve the environmental impact of waste management within the mining sector. To investigate the effects of using lead–zinc tailings as a partial replacement for ordinary Portland cement (OPC), an experimental study was conducted. Two types of lead–zinc tailings were utilized in the experiments to replace 10% and 20% of OPC. A mechanical activation method was adopted using a vibratory cup mill. The effects of activation on the tailings’ particle size distributions and mineralogy were evaluated. The results indicated that the activation was insufficient to promote the pozzolanic activity in T1 and only partially promoted it in T2. A total of 18 different tailing-based mortar (TBM) specimens were produced from the raw and activated tailings, and their flowability, setting time, and compressive strengths after 7, 28, and 90 days were evaluated. The microstructures of the specimens were analyzed using scanning electron microscopy with energy dispersive X-ray spectroscopy. No alteration of mineralogy was observed in T1 after activation; however, a reduction in muscovite was observed in T2. The TBM specimens with 10% activated tailings exhibited comparable 28-day compressive strengths to the control specimen. For the replacement level above 10%, there was a loss of compressive strength at 28 days, both for the activated and raw tailings and for both T1 and T2. Evaluation of the microstructure showed that the use of tailings caused regions in the cement matrix with high metal concentrations. Microcracks could be observed in or around such grains in several cases. The study demonstrated that 10% of OPC can be replaced by lead–zinc tailings while retaining the compressive strength of the specimens. Full article

Accurate, nondestructive assessment of fresh tea leaf quality is important for breeding and field management, yet most spectral work still targets processed or low-moisture products. Here, a mechanistically guided hyperspectral method was developed to estimate free amino acids (AA) and total polyphenols (TP) in fresh leaves. Spectral experiments on purified AA and TP powders and their water mixtures identified a key spectral window at 1660 nm. Fractional-order derivatives were applied to leaf reflectance spectra from 102 spring samples (53 varieties), and full-spectrum Partial Least Squares Regression (PLSR) models were used as comparison and validated on an independent set of 40 summer samples. PLSR achieved decent cross-validation coefficient of determination accuracy for AA ($R_{\mathrm{c}\mathrm{v}}^2=0.867$) and TP ($R_{\mathrm{c}\mathrm{v}}^2=0.755$) and good external prediction coefficient of determination accuracy ($R_{\mathrm{P}}^2=0.793$ and $0.776$, respectively). Guided by the powder and leaf-level analysis, two-band NDSI indices were derived: the AA index of 1735/1626 nm (R²ₚ = 0.687, RPDₚ = 1.788) and the TP index of 1673/1660 nm (R²ₚ = 0.785, RPDₚ = 2.157) approached the PLSR, indicating that much of the useful information for AA and TP is concentrated in this narrow window and can be captured by simple, interpretable indices potentially suitable for in-field sensing, pending validation across multiple sites, seasons, and management conditions. Full article

Background and objectives: Non-surgical deaths in the Emergency Department (ED) occur in the context of severe acute pathology and frequently under conditions of limited diagnostic time and incomplete clinical information. Data integrating ante-mortem clinical assessment with medico-legal autopsy results remain scarce, particularly in Central and Eastern Europe. Materials and Methods: We conducted a retrospective, monocentric descriptive clinicopathological study including 45 consecutive non-surgical deaths occurring in the Emergency Department of a tertiary care hospital between January 2019 and December 2023. Clinical, biological, and temporal data were retrospectively analyzed and correlated with complete medico-legal autopsy findings in order to establish the cause of death and to assess clinicopathological concordance. Results: The mean patient age was 74.3 years, and the median time from ED admission to death was 142 min. Cardiovascular disease was the most frequent cause of death in this cohort (35.6%), followed by sepsis (22.2%), non-COVID respiratory causes (15.6%), and SARS-CoV-2 infection (17.8%). Complete clinicopathological concordance was observed in 37.8% of cases, while partial concordance predominated (57.8%). Total discordance was rare (4.4%). Autopsy findings frequently demonstrated multisystem involvement, particularly in deaths attributed to sepsis and COVID-19. Conclusions: In this descriptive, autopsy-based cohort, non-surgical deaths in the Emergency Department were associated with advanced disease severity and rapid clinical deterioration, limiting complete etiological clarification prior to death. The high rate of partial clinicopathological concordance may reflect the complexity of terminal pathophysiological mechanisms encountered in emergency settings. Systematic clinicopathological correlation through autopsy remains essential for understanding selected cases of acute non-surgical mortality in selected, rapidly fatal ED cases. Full article

Table question answering has been studied using datasets drawn from a variety of tabular sources and task formats. However, most publicly available resources have been created in high-resource languages such as English. For low-resource languages, researchers are often required to construct new datasets or translate existing ones, which incurs substantial time, effort, and financial cost. In contrast to natural language text, table data consists of structured entries whose interpretation is less affected by language-specific syntax or word order. In this work, we present a cost-effective strategy for multilingual table QA that relies on selectively translating only the questions of existing datasets. Leveraging the language-agnostic structure of tables, our approach maintains the original table content while translating queries into multiple target languages. To address possible performance drops caused by using table data in the source language rather than the target language, we apply cross-lingual adaptation techniques using contrastive learning and adversarial training. In addition, to strengthen reasoning ability while avoiding degradation in languages not seen during pre-training, we perform supplementary pre-training of a RoBERTa-based multilingual encoder with SQL-derived table data. Finally, we extend our investigation beyond encoder-based architectures and evaluate decoder-only large language models under the same multilingual table QA setting. The experiments show that LLaMA-3 models exhibit strong cross-lingual generalization even without using translated table context and often achieve competitive performance using only Korean table data. Moreover, the performance gap among training configurations such as translated queries or translated datasets is notably smaller compared to encoder-based models, highlighting the inherent multilingual robustness of modern LLMs. We further evaluate LLaMA-3 models on domain-specific table datasets and observe that domain knowledge acquired from Korean tables transfers effectively across languages even without multilingual supervision, underscoring the potential of LLMs for specialized multilingual table reasoning. These findings demonstrate that LLMs can serve as an effective alternative for multilingual table QA, particularly in low-resource or partially translated environments. Full article

Leaf mass per area (LMA) plays an important role in vegetation productivity, carbon cycling, and remote sensing-based ecosystem monitoring. However, remotely predicting LMA from hyperspectral reflectance remains challenging due to the weak and strongly overlapping spectral response of LMA and spectral variability across species. To address these limitations, this study proposed an integrated framework that combines a fractional-order spectral derivative (FOD) with a one-dimensional convolutional neural network (1D-CNN) to enhance LMA prediction accuracy and cross-species generalization. Leaf hyperspectral reflectance was processed using FOD with 0–2 orders, and the relationship between FOD-enhanced spectra and LMA was analyzed. Model performance was assessed using (i) overall prediction accuracy by an 8:2 random split between training and test sets, and (ii) cross-species generalization through leave-one-species-out validation. The results demonstrated that the 1D-CNN using a 1.5-order derivative achieved the best performance (R² = 0.85; RMSE = 11.57 g/m²), outperforming common machine-learning models including partial least squares regression (PLSR), random forest (RF), and support vector regression (SVR). The proposed method also demonstrated great generalization in cross-species prediction. These results indicate that integrating FOD with 1D-CNN effectively enhances LMA-related spectral information and improves LMA prediction across various species. It provides a promising pathway for applying airborne and satellite hyperspectral images in vegetation biochemical parameter mapping, crop monitoring, and ecological assessment. Full article

In this work, we develop a disturbance suppression-oriented fuzzy sliding mode secured sampled-data controller for third-order parabolic partial differential equations that ought to cope with nonlinearities, hybrid cyber attacks, and modeled disturbances. This endeavor is mainly driven by formulating an observer model with a T–S fuzzy mode of execution that retrieves the latent state variables of the perceived system. Progressing onward, the disturbance observers are formulated to estimate the modeled disturbances emerging from the exogenous systems. In due course, the information received from the system and disturbance estimators, coupled with the sliding surface, is compiled to fabricate the developed controller. Furthermore, in the realm of security, hybrid cyber attacks are scrutinized through the use of stochastic variables that abide by the Bernoulli distributed white sequence, which combat their unpredictability. Proceeding further in this framework, a set of linear matrix inequality conditions is established that relies on the Lyapunov stability theory. Precisely, the refined looped Lyapunov–Krasovskii functional paradigm, which reflects in the sampling period that is intricately split into non-uniform intervals by leveraging a fractional-order parameter, is deployed. In line with this pursuit, a strictly $({\Phi }_1,{\Phi }_2,{\Phi }_3)-\varrho$ dissipative framework is crafted with the intent to curb norm-bounded disturbances. A simulation-backed numerical example is unveiled in the closing segment to underscore the potency and efficacy of the developed control design technique. Full article

For the first time, this paper introduces Nabla fractional calculus into the distributed Nash equilibrium (NE) seeking problem of aggregative games (AGs) with partial decision information in undirected communication networks, and proposes two novel fractional-order distributed algorithms. In the considered setting, each agent can access to only local information and collaboratively estimates the global aggregate through communication with its neighbors. Both algorithms adopt a backward-difference scheme followed by an implicit fractional-order gradient descent step. One updates local aggregate estimates via fractional-order dynamic tracking and the other uses fractional-order average dynamic consensus protocols. Under standard assumptions, convergence of both algorithms to the NE is rigorously proved using nabla fractional-order Lyapunov stability theory, achieving a Mittag-Leffler convergence rate. The feasibility of the developed schemes is verified via numerical experiments applied to a Nash-Cournot game and the coordination control of flexible robotic arms. Full article

Instruction Set Architecture (ISA) extensions, particularly scalar cryptography extensions (Zk), combine the performance advantages of hardware with the adaptability of software, enabling the direct and efficient execution of cryptographic functions within the processor pipeline. This integration eliminates the need to communicate with external cores, substantially reducing latency, power consumption, and hardware overhead, making it especially suitable for embedded systems with constrained resources. However, current scalar cryptography extension implementations remain vulnerable to physical threats, notably power side-channel attacks (PSCAs). These attacks allow adversaries to extract confidential information, such as secret keys, by analyzing the power consumption patterns of the hardware during operation. This paper presents an optimized and secure implementation of the RISC-V scalar Advanced Encryption Standard (AES) extension (Zkne/Zknd) using Domain-Oriented Masking (DOM) to mitigate first-order PSCAs. Our approach features optimized assembly implementations for partial rounds and key scheduling alongside pipeline-aware microarchitecture optimizations. We evaluated the security and performance of the proposed design using the Xilinx Artix7 FPGA platform. The results indicate that our design is side-channel-resistant while adding a very low area overhead of 0.39% to the full 32-bit CV32E40S RISC-V processor. Moreover, the performance overhead is zero when the extension-related instructions are properly scheduled. Full article

This study aims to investigate the tip resistance mechanism of open-ended steel pipe piles under partially plugged conditions by decomposing the load-sharing contribution of the ring zone and the internal soil core. A virtual static loading test was performed using the two-dimensional discrete element method (2D-DEM). Note that the findings of this study were obtained within the range of the 2D-DEM analysis conditions and do not intend to directly reproduce the three-dimensional arching mechanism or to establish equivalence between 2D and 3D responses. Quasi-static conditions were ensured by identifying loading parameters such that the energy residual remained ≤5% during driving, rest, and static loading phases, and the sensitivity criterion |Δq_b|/q_b ≤ 3% was satisfied when the loading rate was halved or doubled. The primary evaluation range of static loading was set to s/D = 0.1 (10% D), corresponding to the displacement criterion for confirming the tip resistance in the Japanese design specifications for highway bridges. For reference, the post-peak mechanism was additionally tracked up to s/D = 0.2 (20% D). Within a fixed evaluation window located immediately beneath the pile tip, high-contact-force (HCF) points were binarized using the threshold τ = μ + σ, and their occupancy ratio φ and normalized force intensity I* were calculated separately for the ring and core regions. A density-based contribution index (“K_{-density share}”) was defined by combining “strength × area” and normalizing by the geometric width. The results suggest that, for the sand conditions and particle-scale ratios examined (D/d_50 = 25–100), the ring zone tends to carry on the order of 85–90% of the tip resistance within the observed cases up to the ultimate state. Even at high plugging ratios (CRs), the internal soil core gradually increases its occupancy and intensity with settlement; however, high-contact-force struts beneath the ring remain active, and it is suggested that the ring-dominant load-transfer mechanism is generally preserved. In the post-peak plastic regime, the K_{-density share} remains around 60%, indicating that the internal core plays a secondary, confining role rather than becoming dominant. These findings suggest that the conventional plug/unplug classification based on PLR can be supplemented by a combined use of plugging ratio CR (a kinematic indicator) and the ring contribution index (K_{-density share}), potentially enabling a continuous interpretation of plugged and unplugged behaviors and contributing to the establishment of a design backbone for tip resistance evaluation. Calibration of design coefficients, scale regression, and mapping to practical indices such as N-values will be addressed in part II of this study. (Note: “Contribution” in this study refers to the HCF-based density contribution index K_{-density share}, not the reaction–force ratio.) Full article

SHapley Additive exPlanations (SHAP) analysis has been applied in disease diagnosis and treatment effect evaluation. However, its application in the prediction and diagnosis of dairy cow diseases remains limited. We investigated whether the variance and autocorrelation of deviations in daily activity, rumination time, and milk electrical conductivity, along with daily milk yield, could be used to predict clinical mastitis in dairy cows using popular machine learning (ML) algorithms and identifying key predictive features using SHAP analysis. Quantile regression (QR) with second- or third-order polynomial models with the median or upper quantiles was used to process raw data from mastitic and healthy cows. Nine variables from the 14-day period preceding mastitis onset were identified as significantly associated with mastitis through logistic regression. These variables were used to train and validate prediction models using eleven classical ML algorithms. Among them, the partial least squares model demonstrated superior performance, achieving an AUC of 0.789, sensitivity of 0.500, specificity of 0.947, accuracy of 0.793, precision of 0.833, and F1-score of 0.625. SHAP analysis results revealed positive contributions of three features to mastitis prediction, whereas two features had negative contributions. These findings provide a theoretical basis for developing clinical decision-support tools in commercial farming settings. Full article

This paper is devoted to the investigation of the proper strict efficient solutions to a set optimization problem with a partial set order relation. Firstly, the notion of proper strict efficient solution defined by the Minkowski difference is introduced, and it is worth mentioning that the introduced strict efficiency is different from those in the existing literature. Secondly, a class of generalized contingent derivatives for set-valued maps is proposed, which are characterized in terms of a set criterion. Finally, the necessary and sufficient optimality conditions and a scalarization theorem for proper strict efficiency are established. Some concrete examples are given to illustrate the obtained results. Full article

The European Water Framework Directive (2000/60/EC) (WFD) aims to protect inland surface waters, transitional waters, coastal waters and groundwater. The overarching goal of the WFD is to reach a good aquatic ecosystem throughout all of Europe. With the aim of reaching this goal, article 4 of the WFD sets certain environmental objectives. According to article 4 of the WFD, all the surface water bodies falling under its scope should be of good chemical and ecological quality by the end of 2027, as most extension deadlines will expire. For artificial and heavily modified surface water bodies—which make up the vast majority in the Netherlands—the goal is not to achieve a good ecological status but instead a good ecological potential and a good chemical status. Point source discharges may have a major impact on water quality and in order to progress, a well-functioning permitting, supervision and enforcement (PSE) process is of considerable interest. So far little academic attention has been paid to the functioning and quality of the governance processes underlying the PSE process. This paper aims to reduce this knowledge gap by conducting a case study on Sitech, the wastewater company for the Chemelot industrial complex in Geleen in the province of Limburg, the Netherlands. We first elaborate on an assessment framework consisting of 18 good governance criteria. The framework is used to assess the permitting, supervision and enforcement process concerning the discharges of Chemelot industrial plant. Our assessment reveals that, despite significant improvements over the last decade, good governance in this case is only partially achieved. While in terms of accountability and resilience the process shows robust strengths, gaps are found in its inclusiveness, effectiveness and transparency. We conclude our paper with some reflections on the generalizability of our findings and some suggestions for further research and policymaking. Full article

This article defines an inverted Topp–Leone distribution. Several mathematical properties and maximum likelihood estimation of parameters of this distribution are considered. The shape of the distribution for different sets of parameters is discussed. Several mathematical properties such as the cumulative distribution function, mode, moment-generating function, survival function, hazard rate function, stress-strength reliability R, moments, Rényi entropy, Shannon entropy, Fisher information matrix, and partial ordering associated with this distribution, have been derived. Distributions of the sum and quotient of two independent inverted Topp–Leone variables have also been obtained. Full article

Over the past thirty years, the focus in singular optics has been on structured beams carrying orbital angular momentum (OAM) for diverse applications in science and technology. However, as practice has shown, the OAM-free structured Gaussian beams with several degrees of freedom are no worse than the OAM beams, especially when propagating through turbulent flows. In this paper, we partially fillthis gap by theoretically and experimentally mapping cyclic changes in vortex-free states (including OAM) as a phase portrait of the beam evolution in an astigmatic optical system. We show that those cyclic variations in the beam parameters are accompanied by the accumulation of the geometric Berry phase, which is an additional degree of freedom. We find also that the geometric phase of cyclic changes in the intensity ellipse shape does not depend on the radial numbers of the Laguerre–Gaussian mode with zero topological charge and is always set by changing the shape of the Gaussian beam. The Stokes parameter formalism was developed to map the beam states’ evolution onto a Poincaré sphere based on physically measurable second-order intensity moments. Theory and experiment are found to be in a good enough agreement. Full article