Search

Herein, the experimental results of discontinuous ferromagnetic‐fiber manipulation using magnetic‐field interaction are presented. High‐aspect‐ratio steel fibers are realigned in viscous polydimethylsiloxanes (PDMSs) with electromagnets. The influence of the matrix viscosity, magnetic‐field strength, and fiber‐aspect ratio on the dynamics of the realignment process is investigated. To predict the fiber behavior under static magnetic fields, two commonly used theoretical models (hereinafter referred to as the torque model and the exponential model) are compared to a nonlinear regression model proposed by the authors. The latter model focuses on the alignment rate of a fiber as a function of the physical process parameters. To indicate the best fitting model for the experimental results, the statistical root‐mean‐square error (RMSE) values of the fitted models are assessed. A total of 54 comparisons are made between the exponential model and the authors' nonlinear regression model. In 50 cases, the nonlinear regression model provides a better fit. [ABSTRACT FROM AUTHOR]

For power systems to operate reliably and effectively, state estimate is essential. For monitoring and control applications, it offers real-time estimates of the system's state indicators, including voltage levels and orientations. In this paper, a novel catagonfly optimization (CFO) technique that aims to improve state estimation effectiveness in electrical systems is presented. The suggested strategy combines optimization techniques from cat swarm and dragonfly. The suggested strategy is evaluated using the IEEE-118 test environment under various simulated scenarios, and the findings are contrasted with those of other methods already in use. The optimization results and statistical error assessment demonstrate the suggested CFO technique's efficiency to the alternatives. The results of this study also have the potential to improve the incorporation of sustainable energy sources, microgrids, and additional emerging innovations into existing power grids. [ABSTRACT FROM AUTHOR]

Litchi florescence has large flower spikes and volume; reasonable control of the ratio of male to female litchi flowers is the key operational aspect of litchi orchards for preserving quality and increasing production. To achieve the rapid detection of male and female litchi flowers, reduce manual statistical errors, and meet the demand for accurate fertilizer regulation, an intelligent detection method for male and female litchi flowers suitable for deployment to low-power embedded platforms is proposed. The method uses multi-teacher pre-activation feature distillation (MPFD) and chooses the relatively complex YOLOv4 and YOLOv5-l as the teacher models and the relatively simple YOLOv4-Tiny as the student model. By dynamically learning the intermediate feature knowledge of the different teacher models, the student model can improve its detection performance by meeting the embedded platform application requirements such as low power consumption and real-time performance. The main objectives of this study are as follows: optimize the distillation position before the activation function (pre-activation) to reduce the feature distillation loss; use the LogCosh-Squared function as the distillation distance loss function to improve distillation performance; adopt the margin-activation method to improve the features of the teacher model passed to the student model; and propose to adopt the Convolution and Group Normalization (Conv-GN) structure for the feature transformation of the student model to prevent effective information loss. Moreover, the distilled student model is quantified and ported for deployment to a field-programmable gate array (FPGA)-embedded platform to design and implement a fast, intelligent detection system for male and female litchi flowers. The experimental results show that compared with an undistilled student model, the mAP of the student model obtained after MPFD feature distillation is improved by 4.42 to 94.21%; the size of the detection model ported and deployed to the FPGA-embedded platform is 5.91 MB, and the power consumption is only 10 W, which is 73.85% and 94.54% lower than that of the detection models on the server and PC platforms, respectively, and it can better meet the application requirements of rapid detection and accurate statistics of male and female litchi flowers. [ABSTRACT FROM AUTHOR]

Background & objectives: Due to lack of appropriate statistical knowledge, published research articles contain various errors related to the design, analysis and interpretation of results in the area of biomedical research. If research contains statistical error, however, costly, it may be of no use and the purpose of the investigation gets defeated. Many biomedical research articles published in different peer reviewed journals may retain several statistical errors and flaws in them. This study aimed to examine the trend and status of application of statistics in biomedical research articles. Study design, sample size estimation and statistical measures are crucial components of a study. These points were evaluated in published original research articles to understand the use or misuse of statistical tools. Methods: Three hundred original research articles from the latest issues of selected 37 journals were reviewed. These journals were from the five internationally recognized publication groups (CLINICAL KEY, BMJ Group, WILEY, CAMBRIDGE and OXFORD) accessible through the online library of SGPGI, Lucknow, India. Results: Among articles assessed under present investigation, 85.3 per cent (n=256) were observational, and 14.7 per cent (n=44) were interventional studies. In 93 per cent (n=279) of research articles, sample size estimation was not reproducible. The simple random sampling was encountered rarely in biomedical studies even though none of the articles was adjusted by design effect and, only five articles had used randomized test. The testing of assumption of normality was mentioned in only four studies before applying parametric tests. Interpretation & conclusions: In order to present biomedical research results with reliable and precise estimates based on data, the role of engaging statistical experts need to be appreciated. Journals must have standard rules for reporting study design, sample size and data analysis tools. Careful attention is needed while applying any statistical procedure as, it will not only help readers to trust in the published articles, but also rely on the inferences the published articles draw. [ABSTRACT FROM AUTHOR]

Weak lensing is one of the most powerful probes for dark matter and dark energy science, although it faces increasing challenges in controlling systematic uncertainties as the statistical errors become smaller. The point spread function (PSF) needs to be precisely modeled to avoid systematic error on the weak lensing measurements. The weak lensing biases induced by errors in the PSF model second moments, i.e. its size and shape, are well-studied. However, Zhang et al. showed that errors in the higher moments of the PSF may also be a significant source of systematics for upcoming weak lensing surveys. Therefore, this work comprehensively investigate the modelling quality of PSF moments from the 3rd to 6th order, and propagate the PSFEx higher moments modelling error in the HSC survey data set to the weak lensing shear–shear correlation functions and their cosmological analyses. The overall multiplicative shear bias associated with errors in PSF higher moments can cause an ∼0.1σ shift on the cosmological parameters for LSST Y10, while the associated additive biases can induce 1σ uncertainties in cosmology parameter inference for LSST Y10, if not accounted. We compare the PSFEx model with PSF in Full FOV, and find similar performance in modelling the PSF higher moments. We conclude that PSF higher moment errors of the future PSF models should be reduced from those in current methods, otherwise needed to be explicitly modeled in the weak lensing analysis. [ABSTRACT FROM AUTHOR]

Climate change has increased the severity and frequency of droughts over the last decades. To alleviate the adverse impacts of droughts, an effective planning and management framework requires high-resolution spatiotemporal data. TRMM multi-satellite precipitation analysis (TMPA) dataset provides sufficient accuracy with fine spatio-temporal resolution. However, it only covers a short temporal span, which limits its applicability for drought studies. This paper presents a methodology for efficient and accurate temporal extension of TMPA using four artificial intelligence (AI)-based models. To improve AI-based model precipitation estimations, fusion techniques including Orness, Orlike, and genetic algorithm (GA)-based weighting methods were employed. Results show that fusion approaches provide more accurate estimates of precipitation. Different timescales of n-SPI time series and drought spatial maps were prepared to visually evaluate the performance of long-term TMPA (LT-TMPA) alongside statistical error indices. The results confirm that this dataset is effective for meteorological drought monitoring over southern Iran. Finally, drought risk assessment was carried out to determine the spatiotemporal characteristics of droughts through severity-duration-frequency (SDF) contour maps. In contrast to the traditional SDF curves, SDF contour maps provide a superior understanding of drought for policymakers since they preserve spatial information. [ABSTRACT FROM AUTHOR]

Issues related to using an automated control system for the operation of thermocompression equipment are discussed. The automated system under consideration is designed to control metal flow during the formation of geometrically complex products under the conditions of isothermal and superplastic forging. The identified problems include noncompliance with the temperature and rate parameters of the forging process, the inertia of the system, the complexity of reconfiguring the system when manufacturing parts from different materials or when switching to the use of other equipment. To overcome the inertia of the system, it is proposed to use, on the one hand, digital extrapolation algorithms based on an experimental recording of the motion of operating elements under the influence of electric current, depending on its strength and duration of exposure, and, on the other, pulse–width modulation to control the current supply to the heating elements. Changes were introduced to modernize the hardware of the automated control system of a thermocompression unit and the TermoControl 3.1 Alpha software package. As a result, the updated version TermoControl 3.7 Alpha was created to ensure a higher accuracy in maintaining the temperature and rate parameters of the forging process. The deviation from the specified heating parameters (error) did not exceed 2–4°C; the error of the deformation rate comprised about 0.1·10–3 mm/s. An extensive database of heating, holding, and cooling modes was compiled, based on the results of predicting the operation of various thermocompression equipment types and models. The use of the developed automated control system allows the deviations from the specified modes of heating and deformation to range within the limits of statistical error. [ABSTRACT FROM AUTHOR]

The local spectral width (LSW) of a radio occultation (RO) observation in impact parameter representation is a useful parameter for providing information on the uncertainty associated with the RO bending angle measurement. The LSW can potentially be used to specify the bending angle observation error (BaOE) in the lower troposphere for each individual sounding. This study assesses the usefulness and limitations of LSW in representing BaOE for a global data assimilation system. A two-step scheme is proposed to derive profile-dependent BaOE from LSW. Since the LSW-based BaOE varies with each individual RO observation, it is here designated as a dynamic BaOE (DBOE) in contrast to the traditional statistics-based BaOE specification. A benchmark control run and two sensitivity experiments are conducted with continuous cycling data assimilation using the National Centers for Environmental Prediction (NCEP) Gridpoint Statistical Interpolation (GSI) and Global Forecast System (GFS). The usefulness and impact of the LSW-based DBOE are evaluated using radiosonde observations and global analyses. Results show that DBOE is able to improve the assimilation of RO data, leading to better forecast skill scores. Another experiment, in which the GSI statistical observation error of the benchmark run is replaced by the average of LSW-based DBOE, shows that the ability to assign larger weighting for high-quality observation and lower weighting for low-quality observation is the key factor for the success of the LSW-based DBOE. [ABSTRACT FROM AUTHOR]

To further improve the prediction accuracy of the critical frequency of the ionospheric F2 layer (foF2), we use the machine learning method (ML) to establish an explanatory dynamic model to predict foF2. Firstly, according to the ML modeling process, the three elements of establishing a prediction model of foF2 and four problems to be solved are determined, and the idea and concrete steps of model building are determined. Then the data collection is explained in detail, and according to the modeling process, foF2 dynamic change mapping and its parameters are determined in turn. Finally, the established model is compared with the International Reference Ionospheric model (IRI-2016) and the Asian Regional foF2 Model (ARFM) to verify the validity and reliability. The results show that compared with the IRI-URSI, IRI-CCIR, and ARFM models, the statistical average error of the established model decreased by 0.316 MHz, 0.132 MHz, and 0.007 MHz, respectively. Further, the statistical average relative root-mean-square error decreased by 9.62%, 4.05%, and 0.15%, respectively. [ABSTRACT FROM AUTHOR]

In the present work, polynomial, discrete singular convolution and sinc quadrature techniques are employed as the new techniques to derive accurate and efficient numerical solutions for the reaction–diffusion equations. Three models, Fitzhugh-Nagumo, Newell–Whitehead–Segel, and tumor growth models, were presented. The equations of three models are reduced to nonlinear ordinary differential equations by using different quadrature schemes. Then, Runge–Kutta fourth-order method is employed to solve nonlinear ordinary differential equations. In addition, the MATLAB program is used to solve these problems. Comparisons between the new methods and the existing ones are included, demonstrating the ease of implementation and efficiency. Also, the calculated results are supported by four various statistical errors. It is found that the rate of error reaches ≤ 10–6 in discrete singular convolution depending on regularized Shannon kernel which is better than others. Further, a parametric analysis is presented to discuss the influence of diffusion and reaction parameters on the solution. [ABSTRACT FROM AUTHOR]

Glycemia assessment in people with type 1 diabetes (T1D) has focused on the time spent in different glucose ranges. As this time reflects the relative contributions to the finite duration of a day, it should be treated as compositional data (CoDa) that can be applied to T1D data. Previous works presented a tool for the individual categorization of days and proposed a probabilistic transition model between categories, although validation has hitherto not been presented. In this study, we consider data from eight real adult patients with T1D obtained from continuous glucose monitoring (CGM) sensors and introduce a methodology based on compositional methods to validate the previously presented probability transition model. We conducted 5-fold cross-validation, with both the training and validation data being CoDa vectors, which requires developing new performance metrics. We design new accuracy and precision measures based on statistical error calculations. The results show that the precision for the entire model is higher than 95% in all patients. The use of a probabilistic transition model can help doctors and patients in diabetes treatment management and decision-making. Although the proposed method was tested with CoDa applied to T1D data obtained from CGM, the newly developed accuracy and precision measures apply to any other data or validation based on CoDa. [ABSTRACT FROM AUTHOR]

Although species are being lost at alarming rates, previous research has provided conflicting results on the extent and even direction of global biodiversity change at the local scale. Here, we assessed the ability to detect global biodiversity trends using local species richness and how it is affected by the number of monitoring sites, sampling interval (i.e. time between original survey and re‐survey of the site), measurement error (error of the measurement of the local species richness), spatial grain of monitoring (a proxy for the taxa mobility) and spatial sampling biases (i.e. site‐selection biases). We use PREDICTS model‐based estimates as a proxy for the real‐world distribution of biodiversity and randomly selected monitoring sites to calculate local species richness trends. We found that while a monitoring network with hundreds of sites could detect global change in species richness within a 30‐year period, the number of sites for detecting trends doubled for a decade, increased 10‐fold within three years and yearly trends were undetectable. Measurement errors had a non‐linear effect on statistical power, with a 1% error reducing statistical power by a slight margin and a 5% error drastically reducing the power to reliably detect any trend. The ability to detect global change in local species richness was also related to spatial grain, making it harder to detect trends for sites sampled at smaller plot sizes. Spatial sampling biases not only reduced the ability to detect negative global biodiversity trends but sometimes yielded positive trends. We conclude that detecting accurate global biodiversity trends using local richness may simply be unfeasible with current approaches. We suggest that monitoring a representative network of sites implemented at the national level, combined with models accounting for errors and biases, can help improve our understanding of global biodiversity change. [ABSTRACT FROM AUTHOR]

By exploiting the received power change in a communication link produced by the presence of a human body in an otherwise empty room, this work evaluates indoor free device localization methods in the 28 GHz band using machine learning techniques. For this objective, a database is built using results from ray tracing simulations of a system comprised of 4 receivers and up to 2 transmitters, while a person is standing within the room. Transmitters are equipped with uniform linear arrays that switch their main beams sequentially at 21 angles, whereas the receivers operate with omnidirectional antennas. Statistical localization error reduction of at least 16% over a global-based classification technique can be obtained through the combination of two independent classifiers using one transmitter and a reduction of at least 19% for 2 transmitters. An additional improvement is achieved by combining each independent classifier with a regression algorithm. Results also suggest that the number of examples per class and size of the blocks (strips) in which the study area is partitioned play a role in the localization error. [ABSTRACT FROM AUTHOR]

Summary: In software‐defined networking (SDN), the controller relies on the information collected from the data plane for route planning, load balancing, and other functions. Statistics information is the most important kind of information among them, so the correctness of statistics information is the key to the proper operation of the network. Most of the current research on data plane focuses on policy consistency, rule redundancy, forwarding anomalies, and so on, and little attention is paid to whether the statistics information uploaded by the switches to the controller is correct. However, incorrect statistics information inevitably leads the controller to make wrong decisions. Therefore, this paper proposes an audit‐based malicious information correction mechanism to address the problem of wrong statistics information uploaded by the switches. This mechanism audits the statistics information and locates malicious switches before uploading the statistics information to the controller. It identifies and corrects the statistics information errors by combining flow path and statistics information. We have performed simulations on Nsfnet, Abilene, and Fat‐Tree, and the results show that our method can correct about 70% of the statistical information errors with less computational cost. To the best of our knowledge, this paper is the first malicious statistics information correction scheme for wildcard rules. [ABSTRACT FROM AUTHOR]

Kriging is a prevalent surrogate model technique in optimization and data-driven analysis, known for its high accuracy and statistical error estimation. However, training Kriging models often requires extensive global optimization of hyperparameters, posing significant challenges when applying these methods to large-scale datasets. Previous research has mainly focused on expediting the training process by reducing the dimensions of design variables and hyperparameters, but this approach frequently results in information loss that adversely affects model accuracy. To address this issue, we introduce a novel methodology called KBCC (Kriging utilizing Boxplot and Correlation Coefficient), which: 1) extracts trends and characteristics of the dataset using boxplot parameters and correlation coefficients; 2) quickly estimates the hyperparameters via a pre-trained surrogate model; and 3) finalizes the hyperparameters with a two-step local optimization. In constructing the pre-trained hyperparameter predictor, we utilize datasets from various functions to enhance its prediction capability. This procedure allows for efficient search of optimal hyperparameter combinations without the need for dimension reduction, thereby accelerating Kriging training across diverse datasets while ensuring high accuracy. The performance and efficiency of KBCC are validated through six numerical test problems and three engineering examples, demonstrating its superior performance relative to conventional methodologies. [ABSTRACT FROM AUTHOR]

The objective of this paper is to address and reduce the uncertainties associated with measurement noise and discretization in damage identification methods based on modal curvature analysis. The experimental case study considers the local reduction in the bending stiffness of a slender beam under free-free and simply supported boundary conditions. First, the analysis of error sources and their propagation is theoretically set up. Second, the mitigation of uncertainties in damage localization is pursued using a two-stage approach based on multiple hypothesis testing relative to the normalized indices and the definition of a combined macro-index. Finally, the Monte Carlo method is exploited to obtain the statistical error distribution of the experimental damage position and severity predictions by randomizing the numerical displacement mode shapes with the identified noise. The present analysis allows us to find the optimal number of sensors that minimizes the combination of bias and truncation errors, to highlight how sensor spacing and data noise affect damage localization, and to determine the uncertainty bounds of the predicted damage severity. The two-stage approach, enhanced by selecting thresholds related to real noise levels and tuned on SHM objectives, appears to improve identification accuracy compared to the separate use of damage indices based on absolute confidence levels. • Analysis of error sources and propagation includes discretization, bias and noise on the input modes. • Mitigation of uncertainty in damage localization by multiple hypothesis testing and macro-index definition. • Damage index threshold tailored on the experimental noise levels and accuracy goals in damage detection. • Uncertainty in evaluating the damage location and severity is estimated using Monte Carlo simulations. • Optimal number of sensors minimizing the combination of truncation and bias error is analytically found. [ABSTRACT FROM AUTHOR]

• Model calibration is performed using the ST4 and ST6 physics parameterizations in WW3. • ST4 provides the best model performance for wave conditions off the coast of Chile. • ST6 enhances model performance in simulating wave direction. • A multi-criteria model performance score was defined to select the best model setup. • The upgraded hindcast, AOC3, shows improved results compared to its predecessor, AOC1. The Chilean Wave Atlas (AOC1), a reliable hindcast developed in 2017 for the academic and engineering community, diminished its usefulness due to the obsolescence of wind data provided by ERA-Interim reanalysis. This study presents the calibration and validation of a new wave hindcast forced using hourly winds data from ERA5 reanalysis. A total of 24 simulations were conducted: 15 using the semi-empirical ST4 parameterization and 9 using the observed-based ST6 parameterization, both implemented in WaveWatch III. Model results were compared with in-situ wave data from buoys along the Chilean coast. Generally, the ST4 physics package demonstrated superior performance with minimal variability in error statistical parameters between simulations. However, the observed-based ST6 parameterization produced the best results for simulating wave direction. By defining a multi-criteria performance score, the optimal model configuration was selected, and a new hindcast was generated for the period between 1979 and 2022. This hindcast includes hourly fields of significant wave height, mean and peak wave period, and mean and peak wave direction for the Pacific Ocean, as well as 72 locations with directional spectra. The upgraded Chilean Wave Atlas (AOC3) significantly improves the performance of AOC1 when compared with satellite-derived wave heights along the Chilean coast. Furthermore, the AOC3 data show good performance compared to other freely available hindcasts. [ABSTRACT FROM AUTHOR]

The article "Refractive Outcomes After Scleral Fixation of the Carlevale Intraocular Lens" in the Journal of Refractive Surgery raises concerns about the study's sample size, the use of different devices for measurements, and the potential impact of measurement discrepancies on intraocular lens power calculations. The authors highlight the importance of accurate axial length and keratometry readings in minimizing postoperative errors and emphasize the need for high precision in such calculations. Overall, the authors express significant concerns about the reliability and validity of the study's conclusions. [Extracted from the article]

Subsurface stratigraphic configuration at a site is often challenging to characterize accurately, due to the complexity and inherent spatial variability of strata, along with the limited availability of site investigation data; thus, considerable stratigraphic uncertainty is induced. Site investigation aims to characterize and reduce this uncertainty, whereby borehole and geophysical explorations are the most widely used exploration techniques. However, borehole and geophysical data are often considered as separated information in existing studies. To this end, this paper presents a novel approach for characterizing stratigraphic uncertainty by merging borehole and geophysical data. In this approach, the probability distribution of geophysical parameter corresponding to a specified stratum can be derived based on both data collected at the same locations. Next, the probability of stratum existence can be estimated based upon geophysical data and the evaluated probability distribution of geophysical parameter. Then, potential stratigraphic realizations can be sampled systematically with the conditional random field theory, based upon borehole data and the estimated probability of stratum existence. Two illustrative examples with different stratigraphic structures are analyzed to demonstrate the effectiveness and versatility of the new approach. Further, a series of comparative analyses are performed to reveal the influences of types of site investigation data and their statistical error on site characterization quantitatively, through which the advantages of the fusion of borehole and geophysical data can be illustrated. [ABSTRACT FROM AUTHOR]

Floating offshore wind turbines (FOWTs) are progressing toward deeper water depth. Overcoming the challenges presented by complex marine environments requires the design of mooring systems that balance both cost and performance. This study introduces a new approach to mooring system design integrating Deep Neural Networks (DNN) with the Non-dominated Sorting Genetic Algorithm III (NSGA-III), which optimizes mooring costs, reduces tension in mooring lines, minimizes platform movement, and enhances overall design efficiency. Latin Hypercube Sampling (LHS) explores design possibilities by generating samples within defined parameter limits. Initial static calculations are performed based on catenary theory to narrow down the range of design variables, and these are followed by dynamic simulations using an aero-hydro-mooring coupled model to generate the necessary dataset for analysis. A DNN surrogate model is employed to predict the dynamic behavior of the FOWT, with hyperparameter optimization performed using Bayesian methods. The reliability of this model is confirmed by evaluating its accuracy through five statistical error metrics. NSGA-III is applied to three design load cases to optimize the mooring system, resulting in three corresponding Pareto solution spaces. To improve the relevance of the optimized solutions, intersections between these solution spaces are identified, followed by time-domain analysis to assess their performance across multiple dimensions. The optimized mooring system achieves notable gains in performance and cost-effectiveness compared to the original design. The most economical solution reduces mooring costs by 23.62, the safest configuration increases the safety factor by 25.3, and the least-displacement solution reduces surge by 31.6%. • A mooring system design method using DNN and NSGA-III is proposed. • The method optimizes costs, reduces line tension, and minimizes displacement for a 10 MW FOWT in South China Sea conditions. • The surrogate model significantly reduces computational time with high accuracy. [ABSTRACT FROM AUTHOR]

• The energy consumption of the operator is considered and modeled. • Establish a rig-operator system energy model. • Propose a prediction method of rig EC based on statistical regression and error compensation. • Explore energy-saving strategies by incorporating the interaction between the rig and the operator. • The superiority of the proposed method are verified by experiments. Energy consumption (EC) during drilling and excavation is closely related to the activities of the rig and the operator. However, due to the complexity of the geological environment, the accuracy of EC prediction in rock mass drilling is low, and the EC related to the operator is often ignored. To make up for this gap, this paper analyzes the power flow in the working process of the rig, explores the internal relationship between the working parameters of the rig and the power of the rig in different working stages, establishes the load power model, and calculates the EC value by statistical regression method. The EC prediction error of statistical regression is corrected by the least square support vector machine model optimized by the improved whale algorithm. Next, combined with the energy consumed by the operators, a new energy model of the rig and operator is established. Finally, the results show that the evaluation indexes of the rig EC prediction method proposed are the minimum values under the same conditions, compared with the single statistical regression and data-driven method, the accuracy of the proposed method to predict the EC of the rig is improved by at least 0.27%, and the energy-saving of the rig-operator system is 23.8%, which verifies the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The discretization of chaotic systems introduces statistical errors in addition to discretization errors into the estimation of quantities of interest. In order to efficiently arrive at estimates of quantities of interest, these two forms of error should be balanced; however, simulations are run without knowledge of the true/asymptotic outputs of interest or their error behaviors. In this work, we develop a framework for error modeling and identification using small-sample Bayesian inference that allows approximation of the optimal balance between sampling time and discretization precision without the computation of high-cost libraries of reference solutions. The result enables the possibility of running chaotic and turbulent simulations in a way that minimizes the total error between sampling and discretization without prior knowledge of the error behavior of the system. • For many physical systems, chaotic behavior leads to statistical errors in addition to discretization errors. • Long sampling decreases statistical errors; high simulation fidelity reduces discretization errors; both incur cost. • At fixed cost, statistical and discretization costs must be balanced correctly to achieve the minimum possible error. • We create a model for simulation output and comprehensive error that is compatible with Bayesian inference. • Using Bayesian inference, we demonstrate the ability to make nearly optimal estimates of quantities of interest at low cost. [ABSTRACT FROM AUTHOR]

This paper presents the development and evaluation of the Coarse Mesh Finite Difference (CMFD) acceleration method to The Random Ray Method (TRRM). We demonstrate its effectiveness in accelerating the convergence of the fission sources and reducing the real statistical error in neutron transport calculations. TRRM treats the angular variable of the neutron flux stochastically and performs batch sampling of characteristic rays to transform the Method of Characteristics (MOC) into a stochastic process, that is capable of making significant strides in memory efficiency and computational performance for some problems. Despite its advantages, TRRM exhibits a potential challenge with a large number of inactive cycles and inherent inter-cycle correlation much like the Monte Carlo method. To address this, the CMFD acceleration method is explored and demonstrated to dramatically reduce the number of required inactive cycles and diminish inter-cycle correlation. Results from the numerical analysis of a 2D C5G7 core problem indicate that the application of CMFD leads to enhanced convergence, with the integration of a CMFD acceleration step every cycle offering the most substantial reduction in statistical noise and error. The study reveals that applying CMFD with every cycle effectively resolves the issue of needing inactive cycles and significantly lowers the inter-cycle correlation, thereby providing a more accurate estimation of standard deviation for pin power distributions. We conclude that using CMFD not only minimizes the number of inactive cycles of TRRM – much like normal Monte Carlo transport – but also lowers real statistical error effectively. For a targeted maximum standard deviation of 0.1% in the pin power, the addition of CMFD can decrease the number of necessary active cycles by 41% compared to standard TRRM, as demonstrated by the 2D C5G7 benchmark analysis. [ABSTRACT FROM AUTHOR]

The thermal stability of the bleachable and unbleachable components of the electron spin resonance signals in quartz grains from modern fluvial sediments were investigated. It was found that the unbleachable Al center is thermally more stable than the bleachable component, hence the unbleachable component can be extracted by heating. Protocols for equivalent dose evaluation with preheating are proposed, in which the procedure of bleaching is replaced by "partial annealing" (heating at 280 °C for 15 min) for the Al center and by "annealing" (heating at 400 °C for 1 h) for the Ti-Li and Ti-H centers. The dose recovery test confirmed that the proposed protocol works practically, and that the regenerative protocol is better than the additive dose protocol in the aspect of the statistical errors. Appropriate partial annealing conditions to extract the unbleachable Al centers may need to be investigated for each sample. • The unbleachable Al center can be extracted by heating (partial annealing). • Ti signals regenerated after bleaching and annealing have the same thermal stability. • The protocol replacing bleaching with heating is useful for obtaining an accurate D E. [ABSTRACT FROM AUTHOR]

The TQM Journal retracted an article titled "Achieving organizational performance by integrating industrial Internet of things in the SMEs: a developing country perspective" due to statistical errors. The authors were unable to provide the data for verification, leading to the unreliability of the article's findings. The publisher issued an apology to the readers for any inconvenience caused. [Extracted from the article]

This paper is a contribution in the context of variational data assimilation combined with statistical learning. The framework of data assimilation traditionally uses data collected at sensor locations in order to bring corrections to a numerical model designed using knowledge of the physical system of interest. However, some applications do not have available data at all times, but only during an initial training phase. Hence, we suggest to combine data assimilation with statistical learning methods; namely, deep learning. More precisely, for time steps at which data is unavailable, a surrogate deep learning model runs predictions of the 'true' data which is then assimilated by the new model. In this paper, we also derive a priori error estimates on this statistical variational data assimilation (SVDA) approximation. Finally, we assess the method by numerical test cases. [ABSTRACT FROM AUTHOR]

Predictive solar-desalination models are becoming more widespread using ML and AI. However, forecasting solar still water productivity based on numerous designs still needs to be improved. Herein, we used transfer learning to create precise supervised predictive ANN regression models for water productivity (L/m2.day) predictions based on literature findings. Such observation datasets from single-basin solar stills were utilized to build the random initialization ANN model. The transfer learning method was applied to the latter model by taking the learned network (weights) for fine-tuning the hyperparameters from the earlier developed novel hybrid solar still known as the source (pre-trained) ANN model, to predict the target ANN model. Based on most minor statistical errors, the pre-trained model with 5–64–64-1 architecture and ReLU activation function was the most appropriate for water productivity prediction. All created ANN models were compared to the MLR model. The results revealed that the generated target ANN model outperformed the ANN RI and MLR with OI values of 0.872, 0.834, and 0.803, respectively, in all modeling stages. The target ANN model's accuracy and generalization were sufficient. The target ANN model had residuals of forecasted distillate values of around 1%. This work discusses the significance of transfer learning to generate accurate target ANN models for predicting freshwater outputs in single-slope solar stills, which can be integrated with established theoretically tuned parameters to enhance performance and maximize distillate water yields. • Leveraging Transfer Learning: This study enhances distillate forecasts in solar desalination through predictive ANN regression models, utilizing transfer learning. • Optimal Model Architecture: The pre-trained ANN model (5–64–64-1,ReLU)proves optimal for water prediction, showcasing transfer learning efficacy. • Superior Performance: Comparative analyses highlight the target ANN model over RI ANN and MLR models, with higher OI values and lower residuals. • Robust Generalization: The target ANN model exhibits exceptional accuracy and robust generalization for forecasting freshwater outputs. • Practical Implications: This research aids expense optimization, enhances water production, and impacts solar desalination research and industry. [ABSTRACT FROM AUTHOR]

The 2-m temperature data is a significant indicator for studying the weather extremes and the exchange of water and energy fluxes between the surface and atmosphere. This study compared three reanalysis datasets, i.e., ERA5, ERA5-Land, and MERRA-2, with observations from in-situ sources from 1990 to 2022 using various statistical error metrics and extreme temperature indices over the Arabian Peninsula (AP) region. We selected these reanalysis datasets due to the continuous improvements and higher spatiotemporal output better capturing the temperature variability. The spatiotemporal climatology shows lower temperatures in winter (<15 °C) and maximum in summer (>35 °C); however, the reanalysis data show more deviations in temperature during the cold season than in the warm season. The reanalysis data underestimated the frequency of the cold (<10 °C) and hot (>30 °C) days across the four regions, except ERA5-Land, which closely followed observed data. The strength of the correlation shows better performance (>0.90) in the cold extremes (cold nights and cold days) frequency than the hot extremes. On an interannual scale, reanalysis products exhibit strong correlations (>0.90) with in-situ data across most regions, particularly in winter and autumn, moderate in spring, and weaker in summer. The reanalysis data shows negative biases in the inland regions and positive biases in the coastal areas with consistent root mean square differences (RMSD) spatiotemporally. The differences in performance are due to the topography and poor representation of the energy fluxes, especially in MERRA-2 as well as missing data in observations. This study recommends ERA5-Land as the first choice for extreme weather simulations in the region, followed by MERRA-2 and ERA5 on the same scale, but proper attention is needed when using reanalysis data for cold and hot extremes. • The reanalysis data shows higher temperature variability during cold season than warm season. • The reanalysis data underestimated the frequency of both cold hot extremes across different regions. • The reanalysis data exhibited higher correlation with observed data in simulating cold nights and cold days. • On Interannual scale the winter and spring season exhibited stronger correlation with observations than in the summer and autumn. [ABSTRACT FROM AUTHOR]

The article "Exploring Antecedents for the Circular Economy Capability of Micro, Small and Medium Enterprises: An Empirical Study" published in Business Strategy & the Environment has been retracted due to errors in statistical and mediation analysis. The retraction was agreed upon by the Journal Editor-in-Chief, Richard Welford, ERP Environment, and John Wiley & Sons Ltd. The authors acknowledged inadvertent mistakes, leading to a loss of confidence in the study's results and conclusions. [Extracted from the article]

Objective:The aim of this study is to evaluate the theses of residency in medicine in terms of statistical errors made and thus to contribute to the production of quality scientific publications by ensuring that scientific publishers in the field of medicine are sensitive and careful about statistics when doing their work.Methods:In this study, we investigated 321 thesis theses which are defended from 6 different universities are obtained from the database of the Turkish Higher Education Council. The investigation is was conducted in terms of “Errors Related to p-values”, “Errors Related to Tests”, “Mathematical Notation Errors”, “Statistical Symbol Errors”, “Inappropriate Interpretation”, “Presentation of The the Statistical Method Analysis and Results in The the Incorrect Section of The the Manuscript”, “Errors in Summarizing Data”,“Incomprehensible Statistical Terms” and “Errors in Statistical Terminology”.Results:There was at least one statistical error in all 321 medicine residency theses examined. The most common error was “errors in summarizing data” with a ratio of 70.1% (n=225), while the least common error was “incomprehensible statistical expressions” with a ratio of 14.3% (n=46).Conclusion:As a result, both researchers and consultants who undertake scientific studies have a responsibility to minimize these errors. To prevent statistical errors, students who are doing residency in medicine are required to receive the necessary training in statistical literacy, to have basic statistical knowledge, and to receive consultancy from a biostatistics expert for statistical evaluations. Students who residency in medicine in preventing statistical errors are required to receive the necessary training in statistical literacy, to have basic statistical knowledge, and to receive consultancy from a biostatistics expert for statistical evaluations.

Data analysis is complex due to a myriad of technical problems. Amongst these, missing values and batch effects are endemic. Although many methods have been developed for missing value imputation (MVI) and batch correction respectively, no study has directly considered the confounding impact of MVI on downstream batch correction. This is surprising as missing values are imputed during early pre-processing while batch effects are mitigated during late pre-processing, prior to functional analysis. Unless actively managed, MVI approaches generally ignore the batch covariate, with unknown consequences. We examine this problem by modelling three simple imputation strategies: global (M1), self-batch (M2) and cross-batch (M3) first via simulations, and then corroborated on real proteomics and genomics data. We report that explicit consideration of batch covariates (M2) is important for good outcomes, resulting in enhanced batch correction and lower statistical errors. However, M1 and M3 are error-generating: global and cross-batch averaging may result in batch-effect dilution, with concomitant and irreversible increase in intra-sample noise. This noise is unremovable via batch correction algorithms and produces false positives and negatives. Hence, careless imputation in the presence of non-negligible covariates such as batch effects should be avoided. [ABSTRACT FROM AUTHOR]

Data from 242 ozonesondes launched from ARIES, Nainital (29.40 ∘ N, 79.50 ∘ E; 1793 m elevation), are used to evaluate the Atmospheric Infrared Sounder (AIRS) version 6 ozone profiles and total column ozone during the period 2011–2017 over the central Himalayas. The AIRS ozone products are analysed in terms of retrieval sensitivity, retrieval biases/errors, and ability to retrieve the natural variability in columnar ozone, which has not been done so far from the Himalayan region, having complex topography. For a direct comparison, averaging kernel information is used to account for the sensitivity difference between the AIRS and ozonesonde data. We show that AIRS has more minor differences from ozonesondes in the lower and middle troposphere and stratosphere with nominal underestimations of less than 20 %. However, in the upper troposphere and lower stratosphere (UTLS), we observe a considerable overestimation of the magnitude, as high as 102 %. The weighted statistical error analysis of AIRS ozone shows a higher positive bias and standard deviation in the upper troposphere of about 65 % and 25 %, respectively. Similarly to AIRS, the Infrared Atmospheric Sounding Interferometer (IASI) and the Cross-track Infrared Sounder (CrIS) are also able to produce ozone peak altitudes and gradients successfully. However, the statistical errors are again higher in the UTLS region, which are likely related to larger variability in ozone, lower ozone partial pressure, and inadequate retrieval information on the surface parameters. Furthermore, AIRS fails to capture the monthly variation in the total column ozone, with a strong bimodal variation, unlike unimodal variation seen in ozonesondes and the Ozone Monitoring Instrument (OMI). In contrast, the UTLS and the tropospheric ozone columns are in reasonable agreement. Increases in the ozone values of 5 %–20 % after biomass burning and during events of downward transport are captured well by AIRS. Ozone radiative forcing (RF) derived from total column ozone using ozonesonde data (4.86 mW m -2) matches well with OMI (4.04 mW m -2), while significant RF underestimation is seen in AIRS (2.96 mW m -2). The fragile and complex landscapes of the Himalayas are more sensitive to global climate change, and establishing such biases and error analysis of space-borne sensors will help us study the long-term trends and estimate accurate radiative budgets. [ABSTRACT FROM AUTHOR]

In this paper, we focused on the new stochastic frontier model with the nonlinear autoregressive structure in panel data. In classic finance models, uncorrelated errors are considered, which are frequently not met in empirical situations. In our stochastic frontier model, the composite error consists of two components, the statistical error and technical inefficiency, whereas the technical inefficiency is assumed to be autocorrelated. A semiparametric method is suggested to estimate nonlinear autoregressive function by two steps procedure as parametric Taylor series expansion and nonparametric adjustment factor. For the model parameters estimation, the Expectation-Maximization approach is applied and the performance of the estimation is checked by Monte Carlo simulation. Finally, we investigate the applicability of new model using a real data set. [ABSTRACT FROM AUTHOR]

The loss of podocyte is crucial for diagnosis and prognosis of diabetic kidney disease, whereas commonly two-dimensional methods for quantifying podocyte number existed with issues of low fidelity and accuracy. In this study, clear, unobstructed brain imaging cocktails and computational analysis (CUBIC), one of three-dimensional optical clearing approaches, was used which combines tissue clearing, immunolabeling, and a light-sheet microscope to image and evaluate podocytes in C57BL/6 (C57) and db/db mice. We discovered that 77 podocytes per glomerulus were in C57 mice. On the subject of db/db mice, there were 74 podocytes by the age of 8 w, 72 podocytes by the age of 12 w, and 66 podocytes by the age of 16 w, compared with 76 podocytes in the control group, suggesting that there was a significant decrease in podocyte number in db/db mice with the age of 16 w, showing a trend which positively correlated to the deterioration of kidney function. Sample size estimation using the PASS software revealed that taking 5%, 7.5%, and 10% of the mean podocyte number per glomerulus as the statistical allowable error and 95% as total confidence interval, 33, 15, and 9 glomeruli were independently needed to be sampled in C57 mice to represent the overall glomeruli to calculate podocyte number. Furthermore, in the control group of db/db mice, 36, 18, and 11 glomeruli were needed, compared with 46, 24, and 14 glomeruli in db/db mice by the age of 8 w, 43, 21, and 12 glomeruli by the age of 12 w, and 52, 27, and 16 by the age of 16 w. These findings indicated that precise quantification of podocyte number could judge the progression of diabetic kidney disease. In addition, a small number of glomeruli could be actually representative of the whole sample size, which indicated apparent practicability of CUBIC for clinical use. [ABSTRACT FROM AUTHOR]

We consider the bias introduced by a spatially varying multiplicative shear bias (m -bias) on tomographic cosmic shear angular power spectra. To compute the bias in the power spectra, we estimate the mode-coupling matrix associated with an m -bias map using a computationally efficient pseudo- C ℓ method. This allows us to consider the effect of the m -bias to high ℓ. We then conduct a Fisher matrix analysis to forecast resulting biases in cosmological parameters. For a Euclid -like survey with a spatially varying m -bias, with zero mean and rms of 0.01, we find that parameter biases reach a maximum of |$\sim 10 {{\ \rm per\ cent}}$| of the expected statistical error, if multipoles up to ℓmax = 5000 are included. We conclude that the effect of the spatially varying m -bias may be a subdominant but potentially non-negligible contribution to the error budget in forthcoming weak lensing surveys. We also investigate the dependence of parameter biases on the amplitude and angular scale of spatial variations of the m -bias field, and conclude that requirements should be placed on the rms of spatial variations of the m -bias, in addition to any requirement on the mean value. We find that, for a Euclid -like survey, biases generally exceed |$\sim 30 {{\ \rm per\ cent}}$| of the statistical error for m -bias rms ∼0.02–0.03 and can exceed the statistical error for rms ∼0.04–0.05. This allows requirements to be set on the permissible amplitude of spatial variations of the m -bias that will arise due to systematics in forthcoming weak lensing measurements. [ABSTRACT FROM AUTHOR]

A new efficient ensemble prediction strategy is developed for a multiscale turbulent model framework with emphasis on the nonlinear interactions between large and small-scale variables. The high computational cost in running large ensemble simulations of high-dimensional equations is effectively avoided by adopting a random batch decomposition of the wide spectrum of the fluctuation states, which is a characteristic feature of the multiscale turbulent systems. The time update of each ensemble sample is then only subject to a small portion of the small-scale fluctuation modes in one batch, while the true model dynamics with multiscale coupling is respected by frequent random resampling of the batches at each time updating step. We investigate both theoretical and numerical properties of the proposed method. First, the convergence of statistical errors in the random batch model approximation is shown rigorously independent of the sample size and full dimension of the system. Next, the forecast skill of the computational algorithm is tested on two representative models of turbulent flows exhibiting many key statistical phenomena with a direct link to realistic turbulent systems. The random batch method displays robust performance in capturing a series of crucial statistical features with general interests, including highly non-Gaussian fat-tailed probability distributions and intermittent bursts of instability, while requires a much lower computational cost than the direct ensemble approach. The efficient random batch method also facilitates the development of new strategies in uncertainty quantification and data assimilation for a wide variety of general complex turbulent systems in science and engineering. [ABSTRACT FROM AUTHOR]

In this work, the emulsified liquid membrane (ELM) extraction process was studied as a technique for separating different pollutants from an aqueous solution. The emulsified liquid membrane used consisted of Sorbitan mono-oleate (Span 80) as a surfactant with n-hexane (C6H14) as a diluent; the internal phase used was nitric acid (HNO3). The major constraint in the implementation of the extraction process by an emulsified liquid membrane (ELM) is the stability of the emulsion. However, this study focused first on controlling the stability of the emulsion by optimizing many operational factors, which have a direct impact on the stability of the membrane. Among the important parameters that cause membrane breakage, the surfactant concentration, the emulsification time, and the stirring speed were demonstrated. The optimization results obtained showed that the rupture rate (Tr) decreased until reaching a minimum value of 0.07% at 2% of weight/weight of Span 80 concentration with an emulsification time of 3 min and a stirring speed of 250 rpm. On the other hand, the volume of the inner phase leaking into the outer phase was predicted using an artificial neural network (ANN). The evaluation criteria of the ANN model in terms of statistical coefficient and RMSE error revealed very interesting results and the performance of the model since the statistical coefficients were very high and close to 1 in the four phases (R_training = 0.99724; R_validation = 0.99802; R_test = 0.99852; R_all data = 0.99772), and also, statistical errors of RMSE were minimal (RMSE_training= 0.0378; RMSE_validation = 0.0420; RMSE_test = 0.0509; RMSE_all data = 0.0406). [ABSTRACT FROM AUTHOR]

Copyright of South China Fisheries Science is the property of South China Fisheries Science Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Superionics are fascinating materials displaying both solid- and liquid-like characteristics: as solids, they respond elastically to shear stress; as liquids, they display fast-ion diffusion at normal conditions. In addition to such scientific interest, superionics are technologically relevant for energy, electronics, and sensing applications. Characterizing and understanding their elastic properties is, e.g., urgently needed to address their feasibility as solid-state electrolytes in all-solid-state batteries. However, static approaches to elasticity assume well-defined reference positions around which atoms vibrate, in contrast with the quasi-liquid motion of the mobile ions in fast ionic conductors. Here, we derive the elastic tensors of superionics from ensemble fluctuations in the isobaric-isothermal ensemble, exploiting extensive Car-Parrinello simulations. We apply this approach to paradigmatic Li-ion conductors, and complement with a block analysis to compute statistical errors. Static approaches sampled over the trajectories often overestimate the response, highlighting the importance of a dynamical treatment in determining elastic tensors in superionics. [ABSTRACT FROM AUTHOR]

Featured Application: This manuscript is part of ongoing research on the strengths and limitations of Wald statistical tests taking into account the incorrect specification of distribution of random effects in generalized linear mixed models. Generalized linear mixed models are commonly used in repeated measurement studies and account for the dependence between observations obtained from the same experimental unit. The designs of repeated measurements in which each experimental unit (e.g., subject) is proven in more than one experimental condition are widespread in psychology, neuroscience, medicine, social sciences and agricultural research. Estimation in generalized linear mixed models is often based on the maximum likelihood theory, which assumes that the assumptions about the underlying probability model are correct. These assumptions include the specification of the distribution of random effects. This research study aimed to identify the impact of the incorrect specification of this distribution on the probability of a type I error and the statistical power of the Wald test. This was achieved through a simulation study where different distributions were considered for random effects in generalized linear mixed models with Poisson and negative binomial responses. Evidence of the impact of the incorrect specification was presented in distributions for random effects different from the normal ones. Lognormal was used for random intercepts and bivariate exponential and Tukey for random intercepts and slopes. Lognormal has positive asymmetry and high kurtosis. Exponential has moderate asymmetry and kurtosis, and Tukey has moderate asymmetry and high kurtosis. [ABSTRACT FROM AUTHOR]

In the years following the reproducibility crisis in behavioral sciences, increased attention of the scientific community has been dedicated to the correct application of statistical inference and promotion of open science practices. In the present survey, we contacted psychology researchers, lecturers, and doctoral students from all universities in Slovakia and the Slovak Academy of Sciences via email. Together we received answers from 65 participants. Questions in the survey covered the most common misconceptions about statistical hypothesis testing, as well as awareness, attitudes, and barriers related to the adherence to open science practices. We found a high prevalence of statistical misconceptions, namely related to the interpretation of p-values and interpretation of null results. At the same time, participants indicated mostly positive attitudes towards open science practices, such as data sharing and preregistration, and were highly interested in further training. These results provide an insight into the needs of the Slovak psychology research community. This is an important step in the further dissemination of open science practices and the prevention of common statistical and methodological errors. [ABSTRACT FROM AUTHOR]

We precisely quantify the impact of statistical error in the quality of a numerical approximation to a random matrix eigendecomposition, and under mild conditions, we use this to introduce an optimal numerical tolerance for residual error in spectral decompositions of random matrices. We demonstrate that terminating an eigendecomposition algorithm when the numerical error and statistical error are of the same order results in computational savings with no loss of accuracy. We illustrate the practical consequences of our stopping criterion with an analysis of simulated and real networks. Our theoretical results and real-data examples establish that the tradeoff between statistical and numerical error is of significant importance for subsequent inference. [ABSTRACT FROM AUTHOR]

Multiple-subject network data are fast emerging in recent years, where a separate connectivity matrix is measured over a common set of nodes for each individual subject, along with subject covariates information. In this article, we propose a new generalized matrix response regression model, where the observed network is treated as a matrix-valued response and the subject covariates as predictors. The new model characterizes the population-level connectivity pattern through a low-rank intercept matrix, and the effect of subject covariates through a sparse slope tensor. We develop an efficient alternating gradient descent algorithm for parameter estimation, and establish the nonasymptotic error bound for the actual estimator from the algorithm, which quantifies the interplay between the computational and statistical errors. We further show the strong consistency for graph community recovery, as well as the edge selection consistency. We demonstrate the efficacy of our method through simulations and two brain connectivity studies. for this article are available online. [ABSTRACT FROM AUTHOR]

Two significant pivotal trials are usually required for a new drug approval by a regulatory agency. This standard requirement is known as the two‐trial paradigm. However, several authors have questioned why we need exactly two pivotal trials, what statistical error the regulators are trying to protect against, and potential alternative approaches. Therefore, it is important to investigate these questions to better understand the regulatory decision‐making in the assessment of drugs' effectiveness. It is common that two identically designed trials are run solely to adhere to the two‐trial rule. Previous work showed that combining the data from the two trials into a single trial (one‐trial paradigm) would increase the power while ensuring the same level of type I error protection as the two‐trial paradigm. However, this is true only under a specific scenario and there is little investigation on the type I error protection over the whole null region. In this article, we compare the two paradigms by considering scenarios in which the two trials are conducted in identical or different populations as well as with equal or unequal size. With identical populations, the results show that a single trial provides better type I error protection and higher power. Conversely, with different populations, although the one‐trial rule is more powerful in some cases, it does not always protect against the type I error. Hence, there is the need for appropriate flexibility around the two‐trial paradigm and the appropriate approach should be chosen based on the questions we are interested in. [ABSTRACT FROM AUTHOR]

The transition kernel of a continuous-state-action Markov decision process (MDP) admits a natural tensor structure. This paper proposes a tensor-inspired unsupervised learning method to identify meaningful low-dimensional state and action representations from empirical trajectories. The method exploits the MDP's tensor structure by kernelization, importance sampling and low-Tucker-rank approximation. This method can be further used to cluster states and actions respectively and find the best discrete MDP abstraction. We provide sharp statistical error bounds for tensor concentration and the preservation of diffusion distance after embedding. We further prove that the learned state/action abstractions provide accurate approximations to latent block structures if they exist, enabling function approximation in downstream tasks such as policy evaluation. [ABSTRACT FROM AUTHOR]

As machine learning powered decision-making becomes increasingly important in our daily lives, it is imperative to strive for fairness in the underlying data processing. We propose a pre-processing algorithm for fair data representation via which supervised learning results in estimations of the Pareto frontier between prediction error and statistical disparity. In particular, the present work applies the optimal affine transport to approach the post-processing Wasserstein barycenter characterization of the optimal fair L²-objective supervised learning via a pre-processing data deformation. Furthermore, we show that the Wasserstein geodesics from the conditional (on sensitive information) distributions of the learning outcome to their barycenter characterize the Pareto frontier between L²-loss and the average pairwise Wasserstein distance among sensitive groups on the learning outcome. Numerical simulations underscore the advantages: (1) the pre-processing step is compositive with arbitrary conditional expectation estimation supervised learning methods and unseen data; (2) the fair representation protects the sensitive information by limiting the inference capability of the remaining data with respect to the sensitive data; (3) the optimal affine maps are computationally efficient even for high-dimensional data. [ABSTRACT FROM AUTHOR]