Search

The current parking trajectory planning algorithms based on geometric connections or formulation of optimization problems in automatic parking systems have strict requirements on the starting position, lower planning efficiency and discontinuous curvature of the reference trajectory. In order to solve these problems, a hierarchical planning algorithm which is combined with nonlinear optimization and the improved RRT* algorithm (Rapidly-exploring Random Tree Star) with Reeds-Shepp curve is proposed in this paper. First, the improved RRT*RS algorithm with the rapid repulsion-straddle experiment is designed for enhancing the efficiency of path planning. Second, because of the shortcomings of the Reeds-Shepp curve that can meet the minimum turning radius but not realize the continuous curvature of the path, a nonlinear optimization problem based on convex-set obstacle constraints is formulated and solved. Finally, simulation results show that the proposed parking trajectory planning algorithm in this paper can plan an effective parking trajectory with continuous curvature in different starting positions and multiple parking scenarios. [ABSTRACT FROM AUTHOR]

Level crossing sampling (LCS) is a power-efficient analog-to-digital conversion scheme for spikelike signals that arise in many Internet of Things-enabled automotive and environmental monitoring applications. However, LCS scheme requires a dedicated time-to-digital converter with large dynamic range specifications. In this paper, we present a compressed LCS that exploits the signal sparsity in the time domain. At the compressed sampling stage, a continuous-time ternary encoding scheme converts the amplitude variations into a ternary timing signal that is captured in a digital random sampler. At the reconstruction stage, a low-complexity split-projection least squares (SPLSs) signal reconstruction algorithm is presented. The SPLS splits random projections and utilizes a standard least squares approach that exploits the ternary-valued amplitude distribution. The SPLS algorithm is hardware friendly, can be run in parallel, and incorporates a low-cost k-term approximation scheme for matrix inversion. The SPLS hardware is analyzed, designed, and implemented in FPGA, achieving the highest data throughput and the power efficiency compared with the prior arts. Simulations of the proposed sampler in an automotive collision warning system demonstrate that the proposed compressed LCS can be very power efficient and robust to wireless interference, while achieving an approximately eightfold data volume compression when compared with Nyquist sampling approaches. [ABSTRACT FROM PUBLISHER]

Purpose A multi-load available, response reliable and product-friendly method is in urgent need to diagnose the signs of incipient arcing. This paper aims to propose a novel algorithm that originates the application of correlativity analysis of wavelet high-frequency component in state discrimination and further in arcing detection.Design/methodology/approach The proposed method calculates the correlation coefficient between the extraction by wavelet transform of arcing series current and that of normal, compares it with a predefined threshold and outputs a trip signal when eight qualified arcing half cycles within a period of 0.5 s are detected.Findings Typical appliances are selected in laboratory for arc detection to test the method which carries on independently of impedance type. The algorithm could be optimized to identify arcing for different kinds of loads, including resistive, inductive, capacitive and switching power supply loads, with a same correlation coefficient threshold.Practical implications The arithmetic operations of the method are addition and multiplication, which contribute to efficient data computation and transmission for micro-processor to undertake. The reference optimal sampling rate recommended for the algorithm helps to reduce the processed data volume and shows its promising prospect for portable product development.Originality/value This proposed correlativity analysis of wavelet transform component algorithm could classify the tested signal into two categories, which benefits the discrimination of normal and fault states in condition monitoring. Laboratory tests prove that it works effectively in arc detection for the commonly used impedance types of loads and needs no offline self-learning or training of samples. [ABSTRACT FROM AUTHOR]

Purpose: Drawing on the theory of belongingness, this study scrutinizes the impact of linguistic ostracism on knowledge sharing, knowledge hiding and knowledge hoarding and the moderating role of cultural intelligence (CQ) in a diverse and multi-cultural work setting. Design/methodology/approach: A two-phase sampling of judgmental and random sampling techniques was used to recruit local and foreign workers in the Jordanian service industry. The present study empirically analyzes the sample of 394 employees' responses by applying variance-based structural equation modeling (VB-SEM). Findings: VB-SEM results indicate that linguistic ostracism lessens knowledge sharing behavior and heightens knowledge hiding and hoarding. CQ moderates two of the said associations, specifically by buffering the causal link between linguistic ostracism and knowledge hiding, as well as linguistic ostracism and knowledge hoarding. Consequently, CQ did not moderate the causal link between linguistic ostracism and knowledge sharing. Practical implications: The study's findings can help managers and decision-makers in such workplaces better understand the deleterious effects of linguistic ostracism and how CQ functions as a buffer. The study concludes with theoretical and managerial implications. Originality/value: Very few investigations have been conducted to determine the consequences of linguistic ostracism in today's multi-cultural and diverse work environment. This paper is among the first to unveil the association of linguistic ostracism and CQ with various knowledge management (KM) concepts. [ABSTRACT FROM AUTHOR]

This paper presents a hardware system that implements a factor graph, where messages are sent using an event-based belief propagation algorithm. The system, comprising an FPGA and an application specific integrated circuit (ASIC) chip, can be used to construct a graph with upto 16 output message channels. The ASIC chip with 16 channels is fabricated in a 0.35 um 2P4M CMOS process and occupies $2.16 \times 2.74$ mm2. Each channel dissipates 46 uW. The output analog messages of the channels are encoded through the interspike intervals of the output spike streams or events. The system can be used to implement graphs with arbitrary variable distributions for its inputs and using constraint functions, such as “plus” and “equality”. Using Kullback–Leibler divergence, we show that the measured distributions from the implemented graphs on the hardware show close similarity to the theoretical distributions. [ABSTRACT FROM AUTHOR]

• The paper is the first paper that provides a comprehensive theoretical analysis to the popular over-sampling method, SMOTE. • The theoretical and empirical analyses demonstrate the divergence of SMOTE patterns from the original minority distribution. • Our theoretical analysis applies to any data distribution, and we provide analysis for two examples of distributions. • We study the impact of number of minority patterns, dimensionality, and K of the KNN used in SMOTE, on SMOTE accuracy. • We study the impact of SMOTE in classification performance theoretically and empirically on LDA, KNN, and SVM classifiers. Imbalanced classification problems are often encountered in many applications. The challenge is that there is a minority class that has typically very little data and is often the focus of attention. One approach for handling imbalance is to generate extra data from the minority class, to overcome its shortage of data. The Synthetic Minority over-sampling TEchnique (SMOTE) is one of the dominant methods in the literature that achieves this extra sample generation. It is based on generating examples on the lines connecting a point and one its K -nearest neighbors. This paper presents a theoretical and experimental analysis of the SMOTE method. We explore the accuracy of how faithful it emulates the underlying density. To our knowledge, this is the first mathematical analysis of the SMOTE method. Moreover, we analyze the effect of the different factors on generation accuracy, such as the dimension, size of the training set and the considered number of neighbors K. We also provide a qualitative analysis that examines the factors affecting its accuracy. In addition, we explore the impact of SMOTE on classification boundary, and classification performance. [ABSTRACT FROM AUTHOR]

This document provides abstracts from the September 2024 issue of the scientific journal Arboriculture & Urban Forestry. The abstracts cover a range of topics related to urban forest management, including factors contributing to conflict and collaboration, the accuracy of different sampling methods for street tree inventory, and the role of urban trees in enhancing thermal comfort. Additionally, the document includes information on managing Japanese beetles as a pest. The Ohio buckeye, a tree native to the Midwestern United States, is also discussed, highlighting its unique characteristics, uses, and potential issues such as leaf scorch and leaf blotch caused by fungus. [Extracted from the article]

• We design a double-layered sampling framework for the registration problem that can greatly truncate the time cost compared with traditional random sample consensus. • We find a time-efficient way to further boost the algorithm speed by introducing a new compatibility-based consensus set building technique. • We combine (1) with (2) to present our DANIEL solver that is robust against high outlier ratios, faster than state-of-the-art algorithms, and applicable to real problems. Correspondence-based point cloud registration is a cornerstone in computer vision, robotics, autonomous navigation and remote sensing, which seeks to estimate the best rigid transformation between two point clouds from the point correspondences established over 3D keypoints. Owing to limited robustness and accuracy, current 3D keypoint matching techniques are rather prone to yield outliers, probably in large numbers, thus making robust estimation (outlier rejection) for the registration problem very important. Unfortunately, previously proposed robust registration methods often suffer from high computation cost or insufficient robustness when encountering high (or even extreme) outlier ratios, hardly ideal enough for practical use. In this paper, we present a novel, time-efficient consensus maximization solver, named Double-layered sAmpliNg with consensus maximization based on stratIfied Element-wise compatibiLity (DANIEL), for robust point cloud registration. DANIEL is smartly designed with two layers of random sampling operation, in order to find the inlier subset with the lowest computational cost possible. Specifically, we (i) apply the rigidity constraint to prune raw outliers in the first layer of one-point sampling, (ii) introduce a series of stratified element-wise compatibility tests to conduct rapid compatibility checking between minimal models so as to realize more efficient consensus maximization in the second layer of two-point sampling, and (iii) employ probabilistic termination conditions to ensure the timely return of the final inlier set. By conducting a series of experiments on multiple real datasets, we demonstrate that the proposed solver DANIEL (i) is robust against over 99% of outliers, (ii) is also significantly faster than previous state-of-the-art robust algorithms, and (iii) is fairly practical for use in real-world applications such as object localization and scan matching problems. [ABSTRACT FROM AUTHOR]

• In this work, we examine the impact of the distribution of random weights on the performance of deep stochastic configuration networks (DeepSCNs). • We propose using symmetric zero-centered distributions with adaptive parameters in constructing DeepSCNs to improve generalization performance. • Demonstrate improvement over uniform distribution using 21 benchmark datasets. Stochastic configuration networks (SCNs) are a class of randomized learner models that ensure the universal approximation property, whereby random weights and biases are drawn from the uniform distribution and selected by a supervisory mechanism. This paper looks into the impact of the distribution of random weights on the performance of SCNs. In the light of a fundamental principle in machine learning, that is, a model with smaller parameters holds improved generalization, we recommend using symmetric zero-centered distributions in constructing SCNs to improve the generalization performance. Further, we introduce a scalar in the distributions to make the SCN model adaptively feasible to different datasets. Simulation results are reported for both regression and classification tasks over twenty-one benchmark datasets using SCN. Results are also presented on ten regression datasets using a deep implementation of SCN, known as deep stochastic configuration networks (DeepSCN). [ABSTRACT FROM AUTHOR]

Abstract In machine learning tasks, it is essential for a data set to be partitioned into a training set and a test set in a specific ratio. In this context, the training set is used for learning a model for making predictions on new instances, whereas the test set is used for evaluating the prediction accuracy of a model on new instances. In the context of human learning, a training set can be viewed as learning material that covers knowledge, whereas a test set can be viewed as an exam paper that provides questions for students to answer. In practice, data partitioning has typically been done by randomly selecting 70% instances for training and the rest for testing. In this paper, we argue that random data partitioning is likely to result in the sample representativeness issue, i.e., training and test instances show very dissimilar characteristics leading to the case similar to testing students on material that was not taught. To address the above issue, we propose a subclass-based semi-random data partitioning approach. The experimental results show that the proposed data partitioning approach leads to significant advances in learning performance due to the improvement of sample representativeness. [ABSTRACT FROM AUTHOR]

In this paper, we consider online support vector machine (SVM) classification learning algorithms with uniformly ergodic Markov chain (u.e.M.c.) samples. We establish the bound on the misclassification error of an online SVM classification algorithm with u.e.M.c. samples based on reproducing kernel Hilbert spaces and obtain a satisfactory convergence rate. We also introduce a novel online SVM classification algorithm based on Markov sampling, and present the numerical studies on the learning ability of online SVM classification based on Markov sampling for benchmark repository. The numerical studies show that the learning performance of the online SVM classification algorithm based on Markov sampling is better than that of classical online SVM classification based on random sampling as the size of training samples is larger. [ABSTRACT FROM AUTHOR]

Abstract The task of recovering a sparse signal from its noisy observations can be divided into two stages; recovering the positions of the nonzero entries or support and estimating their values. In this paper, fundamental limits of sparsity pattern recovery is investigated. Although this problem has been addressed in the recent literature, there is still a gap between those results and the exact limits. In this work, first a tight upper bound is derived on the error probability of the exact sparsity pattern recovery when the Maximum Likelihood (ML) decoder is exploited. Next, a sharp sufficient condition for perfect recovery of the support from noisy projections is derived. In their analyses, the authors treat random sampling matrix with standard normal entries and Gaussian noise. It is shown that the sufficient number of observations for perfect support recovery depends on the noise variance, the minimum nonzero entry of the sparse vector, and the sparsity parameter (k) which is in agreement with previous findings. The proposed upper bound is highly close to the exact probability of error, leading to tight sufficient condition which also matches the growth rate of the necessary conditions of the previous established works. More precisely, the proposed sufficient condition and the previously-established necessary condition on the number of samples are both O (k) (linear in k). The results are proved and supported both theoretically and by simulations. Highlights • In this paper, sufficient condition for exact sparsity pattern recovery is derived for Maximum Likelihood decoder. • A new upper bound is derived on the error probability of the exact sparsity pattern recovery. • The derived upper bound is tighter than the previous upper bound. • Convergence rate of the proposed upper bound is greater than that of the previous bounds. • Results are supported by simulations which show the efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

During the past decade, graph-based learning methods have proved to be an effective tool to make full use of both labeled and unlabeled data samples to improve learning performance. These methods try to discover the intrinsic structures and discriminative information embedded in the data, by building one or more graphs to model the relationship among the data samples. Consequently, how to build an effective graph is the core problem. In this paper we introduce a novel graph-based classification method, called Supervised clustering-based Regularized Least Squares Classification (SuperRLSC), in which local and global graphs of the data are built by supervised clustering. The motivation is that supervised clustering may discover more actual data structures compared to unsupervised clustering. In our algorithm, we firstly employ supervised k-means to partition the whole training dataset into several meaningful clusters in order to discover the intrinsic and discriminative structures. We then use the discovered structures to build local and global graphs of the data. The local graph reveals local geometric and discriminative structures, while the global graph reveals global discriminative information. Finally a hybrid local/global graph-based regularization term is embedded into supervised classification (i.e., RLSC in this paper). To validate the effectiveness of our algorithm, a series of experiments are performed on several UCI benchmark datasets. The results show that our algorithm can achieve better or at least comparable performance to the other graph-based algorithms and the traditional state-of-the-art supervised classification methods. [ABSTRACT FROM AUTHOR]

The proper choice of collective variables (CVs) is central to biased-sampling free energy reconstruction methods in molecular dynamics simulations. The PLUMED 2 library, for instance, provides several sophisticated CV choices, implemented in a C++ framework; however, developing new CVs is still time consuming due to the need to provide code for the analytical derivatives of all functions with respect to atomic coordinates. We present two solutions to this problem, namely (a) symbolic differentiation and code generation, and (b) automatic code differentiation, in both cases leveraging open-source libraries (SymPy and Stan Math, respectively). The two approaches are demonstrated and discussed in detail implementing a realistic example CV, the local radius of curvature of a polymer. Users may use the code as a template to streamline the implementation of their own CVs using high-level constructs and automatic gradient computation. Program summary Program Title: Practical approaches to the differentiation of collective variables in free energy codes: computer-algebra code generation and automatic differentiation Program Files doi: http://dx.doi.org/10.17632/r4r67bvkdn.1 Licensing provisions: GNU Lesser General Public License Version 3 (LGPL-3) Programming languages: C++, Python Nature of problem: The C++ implementation of collective variables (CVs, functions of atomic coordinates to be used in biased sampling applications) in biasing libraries for atomistic simulations, such as PLUMED [1], requires computation of both the variable and its gradient with respect to the atomic coordinates; coding and testing the analytical derivatives complicate the implementation of new CVs. Solution method: The paper shows two approaches to automate the computation of CV gradients, namely, symbolic differentiation with code generation and automatic code differentiation, demonstrating their implementation entirely with open-source software (respectively, SymPy and the Stan Math Library). Additional comments: The paper’s accompanying code serves as an example and template for the methods described in the paper; it is distributed as the two modules curvature_codegen and curvature_autodiff integrated in PLUMED 2’s source tree; the latest version is available at https://github.com/tonigi/plumed2-automatic-gradients . [1] Tribello GA, Bonomi M, Branduardi D, Camilloni C, Bussi G. PLUMED 2: New feathers for an old bird. Computer Physics Communications. 2014 Feb;185(2):604-13. [ABSTRACT FROM AUTHOR]

Statistical matching aims at combining information available in distinct sample surveys referred to the same target population. The matching is usually based on a set of common variables shared by the available data sources. For matching purposes just a subset of all the common variables should be chosen, the so called matching variables. The paper presents a novel method for selecting the matching variables based on the analysis of the uncertainty characterizing the matching framework. The uncertainty is caused by unavailability of data for estimating parameters describing the association between variables not jointly observed in a single data source. The paper focuses on the case of categorical variables and presents a sequential procedure for identifying the most effective subset of common variables in reducing the overall uncertainty. [ABSTRACT FROM AUTHOR]

Objective. To determine the relative risk of a lethal outcome associated with chronic degenerative conditions in patients with COVID-19. Methods. A cohort study was conducted using electronic medical records belonging to patients who tested positive for COVID-19 on RT-PCR while receiving care as outpatients or inpatients in a social security system facility between March 2020 and March 2021. Two study groups were formed. The exposed group was divided into four subgroups, each of which was diagnosed with one and only one chronic condition (diabetes, hypertension, obesity, or chronic kidney disease); the unexposed group was obtained from the medical records of patients without comorbidities. A total of 1 114 medical records were examined using simple random sampling. Once the minimum sample size was reached, the relative risk was calculated for each chronic condition. Combinations of two, three, and four conditions were created, and each of them was included in the analysis. Results. In the absence of a chronic degenerative condition, the prevalence of a lethal outcome from COVID-19 is 3.8%; in the presence of type 2 diabetes mellitus, 15.8%; in the presence of arterial hypertension, 15.6%; and in the presence of obesity, 15.0%. For diabetes and hypertension combined, the prevalence of a lethal outcome is 54.1%; for diabetes and obesity combined, 36.8%, and for obesity and hypertension combined, 28.1%. Conclusion. In patients with COVID-19, the relative risk of a lethal outcome is 4.17 for those with diabetes, 4.13 for those with hypertension, and 3.96 for those with obesity. For two chronic conditions combined, the relative risk doubles or triples. The relative risk of a lethal outcome is 14.27 for diabetes plus hypertension; 9.73 for diabetes plus obesity, and 7.43 for obesity plus hypertension. Chronic conditions do not present alone; they generally occur together, hence the significance of the relative risks for lethal outcomes presented in this paper. [ABSTRACT FROM AUTHOR]

Evaluating similarities between node pairs in a graph is an important task for data analytics and mining. Among various similarity measures proposed in recent years, SimRank is regarded as one of the most influential measures. However, the computation of SimRank is very expensive especially for large graphs. Although pruning technique and random walk based methods were proposed to accelerate the computation, the accuracy of SimRank score is still very low. In this paper, we propose a novel matrix random sampling approach to accelerate computation speed and reduce memory cost. The matrix random sampling technique not only guarantees the sparsity of the involved matrices, but also enhances the precision of estimated SimRank scores. Moreover, we design a fast sparse matrix–matrix multiplication technique which makes the time complexity of single-source query free of the graph size. We further exploit the Steepest Decent technique to accelerate the speed of convergence. The experimental results show our proposed algorithms outperform the state-of-the-art SimRank algorithms. [ABSTRACT FROM AUTHOR]

This paper focuses on investigating the robust stabilization problem for a class of continuous-time networked control systems (NCS) with sampled-data strategy through uncertain time-varying sampling rate via the proportional-integral (PI) control design. The main objective of this paper is to design a state feedback sample-data PI controller with time varying input delay such that the resulting closed-loop system is robustly asymptotically stable for all admissible uncertainties. Based on the Lyapunov stability theory, a new set of delay-dependent conditions is obtained in terms of linear matrix inequalities to determine the system stability and achieve the sampled-data PI control design. Finally, simulation results based on electronic-circuit system and mass–spring damped system are provided to demonstrate the effectiveness and the merit of the proposed design technique. [ABSTRACT FROM AUTHOR]

Efficient computation of skyline probability over uncertain preferences has not received much attention in the database community as compared to skyline probability computation over uncertain data . All known algorithms for probabilistic skyline computation over uncertain preferences attempt to find inexact value of skyline probability by resorting to sampling or to approximation scheme. Exact computation of skyline probability for database with uncertain preferences of moderate size is not possible with any of the existing algorithms. In this paper, we propose an efficient algorithm that can compute skyline probability exactly for reasonably large database. The inclusion–exclusion principle is used to express skyline probability in terms of joint probabilities of all possible combination. In this regard we introduce the concept of zero-contributing set which has zero effect in the signed aggregate of joint probabilities. Our algorithm employs a prefix-based k-level absorption to identify zero-contributing sets. It is shown empirically that only a very small portion of exponential search space remains after level wise application of prefix-based absorption. Thus it becomes possible to compute skyline probability with respect to large datasets. Detailed experimental analysis for real and synthetic datasets are reported to corroborate this claim. We also propose an incremental algorithm to compute skyline probability in dynamic scenarios wherein objects are added incrementally. Moreover, the theoretical concepts developed in this paper help to devise an efficient technique to compute skyline probability of all objects in the database. We show that the exponential search space is pruned once and then for each individual object skyline probability can be derived by inspecting a portion of the pruned lattice. We also use a concept of revival of absorbed pairs . We believe that this process is more efficient than computing the skyline probability individually. [ABSTRACT FROM AUTHOR]

Sequential sampling models are widely used in modeling the empirical data obtained from different decision making experiments. Since 1960s, several instantiations of these models have been proposed. A common assumption among these models is that the subject accumulates noisy information during the time course of a decision. The decision is made when the accumulated information favoring one of the responses reaches a decision boundary. Different models, however, make different assumptions about the information accumulation process and the implementation of the decision boundaries. Comparison among these models has proven to be challenging. In this paper we investigate the relationship between several of these models using a theoretical framework called the inverse first passage time problem. This framework has been used in the literature of applied probability theory in investigating the range of the first passage time distributions that can be produced by a stochastic process. In this paper, we use this framework to prove that any Wiener process model with two time-constant boundaries can be mimicked by an independent race model with time-varying boundaries. We also examine the numerical computation of the mimicking boundaries. We show that the mimicking boundaries of the race model are not symmetric. We then propose an equivalent race model in which the boundaries are symmetric and time-constant but the drift coefficients are time-varying. [ABSTRACT FROM AUTHOR]

In many traditional machine learning methods, sampling is only a process of acquiring training data. However, some studies (on sequential Markov chains and particle filters) have demonstrated that sampling can be used for solving some intractable optimization problems in classical learning methods. Along this line of thinking, the relationships between sampling and learning are theoretically exploited in this paper, wherein the key feature of the sampling process is selecting representative samples from original data that can be modeled by a probability distribution. In theory, acquiring reliable samples is not an easy task for an arbitrary probability distribution. Motivated by approaches in evolutionary computation, rejection sampling and function approximation, a novel sampling strategy, called the evolutionary sampling, is proposed in this paper, and a machine learning method, called the evolutionary sampling approach (ESA), is put forward afterwards. Within ESA, a computing model, called the support sample model (SSM), is presented as well and is used to approximate an original density function. Accordingly, a concrete implementation of an evolutionary sampling approach (ESA) is proposed to seek the optimal model parameters of the SSM. Benefiting from the combination of rejection sampling and evolutionary searching, the ESA can theoretically converge to the optimal solution by minimizing the total variation distance, and can do this with high computational efficiency. Moreover, the normalized factor of a density function can be automatically estimated with high precision within the ESA. As a result, the ESA may be suitable for machine learning problems that could be transformed into density function approximation problems within a probabilistic framework. In addition, derived from the rejection sampling strategy, the ESA can also have online learning abilities required by large-scale data stream processing tasks. Theoretical analyses and application studies are carried out in this paper, and the results demonstrate that the ESA, as a novel way of machine learning, has several prominent merits aspired by past researches in machine learning. [ABSTRACT FROM AUTHOR]

• We proposed the design of a variable length codeword strategy in ECOC encoding algorithm for the first time, which treat the easy classes and hard classes individually by adding more codes for hard classes. • A two-phase strategy is introduced: the first phase generates columns for all classes, and the second phase only tackle hard classes. • The centroids of the two toughest classes are set as the centroids of groups, driving the dichotomizers to concentrate in classifying the hard classes. • Both the random sampling and random feature subspace techniques are utilized in generating training sets for dichotomizers, aiming to enhance the diversity among the ECOC ensemble. • The nearest neighbor search technique is exploited to gather the hard samples, specifically in the hard class stage. At the same time, it reinforces the discriminating power of dichotomizers on the hard classes. Thus far, all existing Error Correcting Output Codes (ECOC) algorithms produce coding matrices with an equal size for all classes. Yet, this paper proposes a variable-length codewords based ECOC (VL-ECOC), which generates longer codes for hard classes than those for easy classes. VL-ECOC consists of two phases: the overall-class phase and the hard-class phase. In the first phase, the centroids of the top two toughest classes are selected as the centroids of the positive group and the negative group respectively, whereas other classes are assigned to their nearer groups. The remaining hard classes with high error rates will be proceeded to the second phase, in which the K nearest neighbors of the misclassified samples are employed to generate new columns. The codewords generated in the second phase are applied to the decoding process of the hard classes. Consequently, both the easy and hard classes contain distinct code lengths. To verify the performance of VL-ECOC, comprehensive experiments are carried out on the UCI data and the microarray data sets. The experiment results demonstrate that owing to the additional codewords for the hard classes, our algorithm can better handle the class imbalance problem and achieve higher performance in most cases. [ABSTRACT FROM AUTHOR]

Retraction statement The publishers of International Journal of Numerical Methods for Heat & Fluid Flow wish to retract the article Khan, N.B., Ibrahim, Z.B., Ali, M.A., Jameel, M., Khan, M.I., Javanmardi, A. and Oyejobi, D.O. (2020), “Numerical simulation of flow with large eddy simulation at Re = 3900: A study on the accuracy of statistical quantities”, International Journal of Numerical Methods for Heat & Fluid Flow, Vol. 30 No. 5, pp. 2397-2409. https://doi.org/10.1108/HFF-11-2018-0619. An internal investigation into a series of submissions has uncovered evidence that the peer review process was compromised. As a result of these concerns, the findings of the article cannot be relied upon. This decision has been taken in accordance with Emerald's publishing ethics and the COPE guidelines on retractions. The authors of this paper would like to note that they do not agree with the content of this notice. The publishers of the journal sincerely apologize to the readers.

The paper studies a variety of domains of preference orders that are closely related to single-peaked preferences. We develop recursive formulas for the number of single-peaked preference profiles and the number of preference profiles that are single-peaked on a circle. The number of Arrow's single-peaked preference profiles is found for three, four, and five alternatives. Random sampling applications are discussed. For restricted tier preference profiles, a forbidden subprofiles characterization and an exact enumeration formula are obtained. It is also shown that each Fishburn's preference profile is single-peaked on a circle preference profile, and Fishburn's preference profiles cannot be characterized by forbidden subprofiles. • The number of single-peaked, single-peaked on a circle, Arrow's single-peaked preference profiles is calculated. • Restricted tier preference profiles are characterized via forbidden subprofiles. • It is shown that Fishburn's domain is a subset of single-peaked on a circle domain. [ABSTRACT FROM AUTHOR]

Knowledge Graph Reasoning (KGR) is an effective method to ameliorate incompleteness and sparsity problems, which infers new knowledge based on existing knowledge. The Graph Neural Network (GNN)-based approaches can obtain advanced effectiveness. However, it still suffers from some problems such as obtaining insufficient graph features, introducing noises, ignoring path connectivity, and acquiring incomplete neighborhood information. This paper proposes a neighborhood-aware (NA) graph self-attention mechanism-based pre-training model for KGR, namely NA-KGR. The proposed model is composed of two phases. The first phase is an enhanced graph attention network, which can use the weighted characteristics of its neighbors to represent and encode the entities that are most likely to have a positive effect on reasoning. The second phase is a neighborhood-aware self-attention mechanism, which makes the model more able to obtain information from the neighbor entities for inference by increasing an adaptive entity similarity matrix when calculating the attention score. Moreover, we propose a pre-training pattern based on neighborhood-aware random walk sampling and general subgraph structure sampling to improve NA-KGR's generalization ability. Extensive comparison and ablation experimental results on various benchmarks unambiguously demonstrate that the proposed NA-KGR model can obtain the state-of-the-art results of current GNN-based methods. [ABSTRACT FROM AUTHOR]

The output-error model structure is often used in practice and its identification is important for analysis of output-error type systems. This paper considers the parameter identification of linear and nonlinear output-error models. A particle filter which approximates the posterior probability density function with a weighted set of discrete random sampling points is utilized to estimate the unmeasurable true process outputs. To improve the convergence rate of the proposed algorithm, the scalar innovations are grouped into an innovation vector, thus more past information can be utilized. The convergence analysis shows that the parameter estimates can converge to their true values. Finally, both linear and nonlinear results are verified by numerical simulation and engineering. [ABSTRACT FROM AUTHOR]

Abstract A crucial task in the process of constructing Internet testbeds is to substantially reduce the number of nodes in the interdomain Internet topology, while maintaining its essential properties. In this paper, we first decompose the Internet topology into seven bipartite graphs, a matching graph and a core graph using two normalized Laplacian spectral features, namely, the weighted spectral distribution and the multiplicity of the eigenvalue 1. Then, we accomplish a marked reduction in size of the topology by sampling and merging these graphs. The proposed approach was tested on realistic Internet graphs that were explored from 2000 to 2016 and a large-scale simulated graph. The experimental results confirm that our approach can effectively reduce the number of nodes in these graphs by more than 96%, while maintaining their essential properties. [ABSTRACT FROM AUTHOR]

Abstract This paper proposes a novel image encryption algorithm based on 3D bit matrix and orthogonal Latin cubes. Unlike most of the existing image encryption algorithms, the proposed algorithm treats the original image as a 3D bit matrix, and uses Latin cubes for confusion and diffusion. The utilization of 3D bit matrix and Latin cubes makes the proposed algorithm achieve not only a desirable level of security, but also high efficiency, so the algorithm is suitable for practical application. [ABSTRACT FROM AUTHOR]

For random samples of size n obtained from m -variate normal distributions, we investigate the classical likelihood ratio tests for their means and covariance matrices in the high-dimensional case. Many researchers analyze the test statistics for the case of n going to infinity and m keeping fixed. In the high-dimensional setting, the objective of this paper is to prove that the likelihood ratio test statistics converge in distribution to normal distributions when both m and n go to infinity with m / n → y ∈ (0 , 1 ]. We obtain this conclusion very intuitively by using Lyapunov's central limit theorem. [ABSTRACT FROM AUTHOR]

This follow-up study assessed the impact of a nitrate-rich diet on salivary nitrate/nitrite levels and the recovery of therapy-induced vascular impairments in a cohort of 39 periodontitis patients treated by standard subgingival mechanical plaque removal (PMPR). At baseline, saliva samples for nitrate/nitrite analysis were collected, and peripheral/central blood and augmentation pressure was documented using the Arteriograph recording system. Immediately after, PMPR vascular parameters were reassessed. All study patients received a randomly allocated supply of a lettuce beverage to be consumed for 14 days, containing either a daily dosage of 200 mg nitrate (test group, n = 20) or being void of nitrate (placebo group, n = 19). At day 14, salivary and vascular parameters were reassessed. Initial salivary and vascular parameters did not differ significantly between the groups. PMPR impaired all vascular parameters in both groups with no differences between the groups. At day 14, salivary nitrate/nitrite levels of the test group were significantly elevated compared to baseline. All vascular parameters had significantly recovered from the impairment inflicted by PMPR. In the placebo group, by contrast, salivary parameters did not differ significantly from baseline, and the recovery of impaired vascular parameters was restricted to a significant improvement of diastolic blood pressure. Correlation analysis identified a significant inverse correlation between salivary nitrate/nitrite sum and central/peripheral blood pressure and augmentation pressure. In conclusion, the data of this subanalysis suggest that increasing salivary nitrate/nitrite levels by a diet rich in nitrate may improve recovery of therapy-induced vascular impairments after PMPR. [ABSTRACT FROM AUTHOR]

This paper proposes a method to detect the sampling rate of discrete time series of diffusion processes. Using the maximum likelihood estimates of the parameters of a diffusion process, we establish a criterion based on the Kullback–Leibler divergence and thereby estimate the sampling rate. Simulation studies are conducted to check whether the method can detect the sampling rates from data and their results show a good performance in the detection. In addition, the method is applied to a financial time series sampled on daily basis and shows the detected sampling rate is different from the conventional rates. [ABSTRACT FROM AUTHOR]

This paper proposes a systematic non-uniform multi-rate estimation and control framework for a periodic event-triggered system which is subject to external disturbance and sensor noise. When the disturbance dynamic model is available, and in order to efficiently estimate the state variable and disturbance from non-uniform slow-rate measurements, a time-varying Kalman filter is designed. When the disturbance dynamic model is not available, a disturbance observer is proposed as an alternative approach. Both the Kalman filter and the disturbance observer are proposed in a non-uniform multi-rate format. Such disturbance estimation enables faster controller updating in spite of slower measurement. Interlacing techniques are used in the control system to uniformly distribute the computational load at each fast sampling instance. Compared to the conventional time-triggered sampling paradigm, the control solution is able to reduce the resource utilization, while maintaining a satisfactory control performance. The proposed control solution will reduce the number of transmissions among devices, which enhances the energy and computational efficiency. Simulation results are provided to validate the effectiveness and benefits of the proposed control algorithms. [ABSTRACT FROM AUTHOR]

Imbalanced datasets are commonly encountered in real-world classification problems. Many machine learning algorithms are originally designed for well-balanced datasets, therefore re-sampling has become an important step to pre-process imbalanced data. This aims to balance the datasets by increasing the samples of the smaller class or decreasing the samples of the larger class, which are known as over-sampling and under-sampling, respectively. In this paper, a sampling strategy that is based on both over-sampling and under-sampling is proposed, in which the new samples of the smaller class are created based on fuzzy logic. Improvement of the datasets is done by the evolutionary computational method of Cross-generational elitist selection, Heterogeneous recombination and Cataclysmic mutation (CHC) that under-samples both the minority and majority samples. Consequently, a hybrid preprocessing method is proposed to re-sample imbalanced datasets. The evaluation is done by applying the Support Vector Machine (SVM), C4.5 decision tree and nearest neighbor rule to train a classification model from the re-sampled training sets. From the experimental results, it can be seen that our proposed method improves both the F − m e a s u r e and AUC. The over-sampling rate and complexity of the classification model are also compared. Our proposed method is found to be superior to all other methods under comparison and it is more robust in different classifiers. [ABSTRACT FROM AUTHOR]

This paper studies the nonparametric estimator of the quantile function under length-biased and right censored data, with the property of length-bias that the residual lifetime share the same distribution as the truncation time. A nonparametric estimator of the quantile function is proposed based on the improved product-limit estimator of distribution function that takes into account the auxiliary information about the length-biased sampling. Asymptotic properties of the estimator are derived, and numerical simulation studies are conducted to assess the performance of the proposed estimator, an application is also given using the Channing house data. [ABSTRACT FROM AUTHOR]

Enterprise credit evaluation model is an important tool for bank and enterprise risk management, but how to construct an effective decision tree (DT) ensemble model for imbalanced enterprise credit evaluation is seldom studied. This paper proposes a new DT ensemble model for imbalanced enterprise credit evaluation based on the synthetic minority over-sampling technique (SMOTE) and the Bagging ensemble learning algorithm with differentiated sampling rates (DSR), which is named as DTE-SBD (Decision Tree Ensemble based on SMOTE, Bagging and DSR). In different times of iteration for base DT classifier training, new positive (high risky) samples are produced to different degrees by SMOTE with DSR, and different numbers of negative (low risky) samples are drawn with replacement by Bagging with DSR. However, in the same time of iteration with certain sampling rate, the training positive samples including the original and the new are of the same number as the drawn training negative samples, and they are combined to train a DT base classifier. Therefore, DTE-SBD can not only dispose the class imbalance problem of enterprise credit evaluation, but also increase the diversity of base classifiers for DT ensemble. Empirical experiment is carried out for 100 times with the financial data of 552 Chinese listed companies, and the performance of imbalanced enterprise credit evaluation is compared among the six models of pure DT, over-sampling DT, over-under-sampling DT, SMOTE DT, Bagging DT, and DTE-SBD. The experimental results indicate that DTE-SBD significantly outperforms the other five models and is effective for imbalanced enterprise credit evaluation. [ABSTRACT FROM AUTHOR]

Jittered Sampling is a refinement of the classical Monte Carlo sampling method. Instead of picking n points randomly from [ 0 , 1 ] 2 , one partitions the unit square into n regions of equal measure and then chooses a point randomly from each partition. Currently, no good rules for how to partition the space are available. In this paper, we present a solution for the special case of subdividing the unit square by a decreasing function into two regions so as to minimize the expected squared L 2 -discrepancy. The optimal partitions are given by a highly nonlinear integral equation for which we determine an approximate solution. In particular, there is a break of symmetry and the optimal partition is not into two sets of equal measure. We hope this stimulates further interest in the construction of good partitions. [ABSTRACT FROM AUTHOR]

The norm of data size estimation is to use uniform random samples whenever possible. There have been tremendous efforts in obtaining uniform random samples using methods such as Metropolis–Hasting random walk or importance sampling [2]. This paper shows that, on the contrary to the common practice, uniform random sampling should be avoided when PPS (probability proportional to size) sampling is available for large data. To develop intuition of the sampling process, we discuss the sampling and estimation problem in the context of graph. The size is the number of nodes in the graph; uniform random sampling corresponds to uniform random node (RN) sampling; and PPS sampling is approximated by random edge (RE) sampling. In this setting, we show that for large graphs RE sampling outperforms RN sampling with a ratio proportional to the normalized graph degree variance. This result is particularly important in the era of big data, when data are typically large and scale-free [3], resulting in large degree variance. We derive the result by giving the variances of RN and RE estimators. Each step of the derivation is supported and demonstrated by simulation studies assuming power law distributions. Then we use 18 real-world networks to verify the result. Furthermore, we show that the performance of random walk (RW) sampling is data dependent and can be significantly worse than RN and RE. More specifically, RW can estimate online social networks but not Web graphs due to the difference of the graph conductance. [ABSTRACT FROM AUTHOR]

This paper addresses the networked H ∞ filtering problem for time-delay systems with quadratically inner-bounded nonlinearities, where the different output sub-vectors of the systems are sampled respectively by sensor groups with different sampling periods (i.e., multirate sampling or MRS). The multirate sampled data are transmitted to a filter through multiple communication channels with network-induced packet dropouts and time-varying transmission delays. Considering MRS’s asynchronous behavior, we propose a matching mechanism to synchronize the successfully transmitted sampled data of the nonlinear system and the sampled data of a filter. Based on the matched sampled data, two types of new continuous-time filters are presented to estimate the objective signal of the nonlinear system. In view of the fact that the lifting technique may fail to model the nonlinear system, the resultant filtering error systems are modeled as continuous-time nonlinear systems with several sawtooth delays. By the Lyapunov-Krasovskii functional approach, some sufficient conditions are obtained to ensure that the filtering error systems are stable with a prescribed H ∞ performance level, and the design methods of the two filters are given. Finally, a numerical example shows the effectiveness of our proposed methods. [ABSTRACT FROM AUTHOR]

Purpose: Telecommuting can reduce traffic congestion, energy consumption, prevalence and a death toll of COVID-19 among employees due to less transportation and fewer physical contacts among employees, on the one hand, and efficiently develop their use of information and communications technology, on the other hand. In this regard, the present study aims to explore antecedents and consequences of telecommuting in public organizations. Design/methodology/approach: The study used a descriptive survey method to collect data. The statistical population includes all employees of government organizations in West Azerbaijan province in 2020, which according to the collected information, their number is equal to 63,079 employees. Based on Cochran's formula, a sample size of 686 people was obtained; stratified random sampling was used to select sampling. The process of calculating the sample volume was such that after referring to the preliminary sample and processing the collected data, the variance of the given answers was approximately 0.446. After obtaining the variance of the data, assuming a maximum acceptable error of 5% and a significance level of 0.05, the Cochran's formula calculated the sample size to be 686 people. In order to collect and measure data for the study, a standard questionnaire and the collected data were analyzed using structural equation modeling. Findings: Findings indicate that there is no meaningful relationship between the employees' physical job conditions or the quality of their life with telecommuting and that telecommuting does not have a significant effect on their life. However, job burnout, training and telecommuting experience have a significant positive effect on telecommuting, which in turn has a positive and significant effect on job security, job flexibility, organizational performance and overall productivity of employees. Research limitations/implications: This research is a cross-sectional study, and its data have been collected in a certain period of time, while longitudinal research can provide a richer result. Future research can benefit from the impact of employee isolation and telecommuter organizational commitment. Originality/value: This study hopes to contribute to the increase of the scientific knowledge in the telecommuting field and to allow organizations to rethink the telecommuting strategies to optimize resources and costs and to improve the organization's productivity without harming the quality of life and well-being of their workers. [ABSTRACT FROM AUTHOR]

Class imbalance is often a problem in various real-world data sets, where one class (i.e. the minority class) contains a small number of data points and the other (i.e. the majority class) contains a large number of data points. It is notably difficult to develop an effective model using current data mining and machine learning algorithms without considering data preprocessing to balance the imbalanced data sets. Random undersampling and oversampling have been used in numerous studies to ensure that the different classes contain the same number of data points. A classifier ensemble (i.e. a structure containing several classifiers) can be trained on several different balanced data sets for later classification purposes. In this paper, we introduce two undersampling strategies in which a clustering technique is used during the data preprocessing step. Specifically, the number of clusters in the majority class is set to be equal to the number of data points in the minority class. The first strategy uses the cluster centers to represent the majority class, whereas the second strategy uses the nearest neighbors of the cluster centers. A further study was conducted to examine the effect on performance of the addition or deletion of 5 to 10 cluster centers in the majority class. The experimental results obtained using 44 small-scale and 2 large-scale data sets revealed that the clustering-based undersampling approach with the second strategy outperformed five state-of-the-art approaches. Specifically, this approach combined with a single multilayer perceptron classifier and C4.5 decision tree classifier ensembles delivered optimal performance over both small- and large-scale data sets. [ABSTRACT FROM AUTHOR]

In this paper a subsampling approach for nonstationary time series with a non-zero mean function is proposed. It is applied for periodically and almost periodically processes. Two statistical tests are constructed. An example with real data is presented. [ABSTRACT FROM AUTHOR]

This paper is concerned with the problem of estimating a population proportion p in a judgment post-stratification (JPS) sampling scheme. Different proportion estimators are considered, among which some are specifically designed to deal with JPS samples with empty strata; and asymptotic normality is established for each. A Monte Carlo simulation study and two examples using data from medical studies are employed to examine the performance of these proportion estimators under both perfect and imperfect ranking and for JPS data both with and without empty strata. It is shown that the JPS scheme improves estimation of the population proportion in a very wide range of settings as compared to simple random sampling (SRS). Also, findings about the relative performance of the different estimators are provided to help practitioners determine which estimator should be used under certain situations. [ABSTRACT FROM AUTHOR]

This paper proposes a novel sampling-based adaptive bounding evolutionary algorithm termed SABEA that is capable of dynamically updating the search space during the evolution process for continuous optimization problems. The proposed SABEA adopts two bounding strategies, namely fitness-based bounding and probabilistic sampling-based bounding, to select a set of individuals over multiple generations and leverage the value information from these individuals to update the search space of a given problem for improving the solution accuracy and search efficiency. To evaluate the performance of this method, SABEA is applied on top of the classic differential evolution (DE) algorithm and a DE variant, and SABEA is compared to a state-of-the-art Distribution-based Adaptive Bounding Genetic Algorithm (DABGA) on a set of 27 selected benchmark functions. The results show that SABEA can be used as a complementary strategy for further enhancing the performance of existing evolutionary algorithms and it also outperforms DABGA. Finally, a practical problem, namely the model calibration for an agent-based simulation, is used to further evaluate SABEA. The results show SABEA’s applicability to diverse problems and its advantages over the traditional genetic algorithm-based calibration method and DABGA. [ABSTRACT FROM AUTHOR]

This paper investigates the problem of network-based leader-following consensus of nonlinear multi-agent systems via distributed impulsive control. First, by taking network-induced delays into account, a nonlinear system with delayed impulses is formulated. Then, a general consensus criterion is derived and several special cases of network-induced delays and network topologies are discussed. Moreover, sufficient conditions on the design of the sampling period, pinned nodes and the coupling strength are provided. The effects of the coupling strength and pinning strategy are further explored for multi-agent systems with an undirected communication graph. Finally, two examples are given to verify the theoretical results. [ABSTRACT FROM AUTHOR]

Statistical depth and related quantile functions, originally introduced for nonparametric description and analysis of multivariate data in a way sensitive to inherent geometry, are in active development for functional data and in this setting offer special options since the data may be visualized regardless of dimension. This paper provides depth-based methods for revealing the structure of a functional data set in terms of relevant sample quantile curves displayed at selected levels, and for constructing and displaying confidence bands for corresponding “population” versions. Also, the usual functional boxplot is enhanced, by adding inner fences to flag possible shape outliers, along with the outer fences that flag location outliers. This enables the boxplot to serve as a stand-alone tool for functional data, as with univariate and multivariate data. Further, the spatial depth approach, well-established for multivariate data, is investigated for nonparametric description of functional data along these lines. In comparison with four other commonly used depth approaches for functional data, over a range of actual and simulated data sets, the spatial depth approach is seen to offer a very competitive combination of robustness, efficiency, computational ease, simplicity, and versatility. [ABSTRACT FROM AUTHOR]

Consider a set of independent random variables with specified distributions or a set of multivariate normal random variables with a product correlation structure. This paper shows how the distributions and moments of these random variables can be calculated conditional on a specified ranking of their values. This can be useful when the ordering of the variables can be determined without observing the actual values of the variables, as in ranked set sampling, for example. Thus, prior information on the distributions and moments from their individual specified distributions can be updated to provide improved posterior information using the known ranking. While these calculations ostensibly involve high dimensional integral expressions, it is shown how the previously developed general recursive integration methodology can be applied to this problem so that they can be evaluated in a straightforward manner as a series of one-dimensional or two-dimensional integral calculations. Furthermore, the proposed methodology possesses a self-correction mechanism in the computation that prevents any serious growth of the errors. Examples illustrate how different kinds of ranking information affect the distributions, expectations, variances, and covariances of the variables, and how they can be employed to solve a decision making problem. [ABSTRACT FROM AUTHOR]

Gaussian process emulators of computationally expensive computer codes provide fast statistical approximations to model physical processes. The training of these surrogates depends on the set of design points chosen to run the simulator. Due to computational cost, such training set is bound to be limited and quantifying the resulting uncertainty in the hyper-parameters of the emulator by uni-modal distributions is likely to induce bias. In order to quantify this uncertainty, this paper proposes a computationally efficient sampler based on an extension of Asymptotically Independent Markov Sampling, a recently developed algorithm for Bayesian inference. Structural uncertainty of the emulator is obtained as a by-product of the Bayesian treatment of the hyper-parameters. Additionally, the user can choose to perform stochastic optimisation to sample from a neighbourhood of the Maximum a Posteriori estimate, even in the presence of multimodality. Model uncertainty is also acknowledged through numerical stabilisation measures by including a nugget term in the formulation of the probability model. The efficiency of the proposed sampler is illustrated in examples where multi-modal distributions are encountered. For the purpose of reproducibility, further development, and use in other applications the code used to generate the examples is freely available for download at https://github.com/agarbuno/paims_codes . [ABSTRACT FROM AUTHOR]

Handling time delays in industrial process control is a major challenge in the dominant pole placement based design of proportional-integral-derivative (PID) controllers due to variable number of zeros and poles which may arise from the Pade approximation of the exponential delay terms in the characteristic polynomials used for stability analysis. This paper proposes a new concept for designing PID controllers with a derivative filter using dominant pole placement method mapped onto the discrete time domain with a suitable choice of the sampling time to convert the continuous time time-delays into finite number of discrete time poles. Here, the continuous-time plant and the filtered PID controller have been discretized using the pole-zero matching method for handling linear dynamical systems, represented by the first order plus time delay with zero (FOPTDZ) transfer function models of the open-loop system under control. We use a swarm intelligence based global optimization method as a sampler to discover the approximate the pattern of the stabilizable region in the controller parameter as well as the design specification space while also satisfying the analytical conditions for pole placement given as higher order polynomials. Simulations on test-bench plants with open-loop stable, unstable, integrating, low-pass, high-pass characteristics have been presented in order to demonstrate the validity and effectiveness of the proposed control design method. • Stabilizable regions of First order plus time delay process with lead are found • The stabilization is based on a random search and optimization method • Time delays are handled by transforming them as equivalent discrete-time system poles • Joint optimization for time-domain performance was carried out satisfying pole-placement polynomial criterion • Stabilization method is benchmarked on a test-bench of stable, unstable, integrating, low/high-pass plants [ABSTRACT FROM AUTHOR]