Search

The landing gear structure suffers from large loads during aircraft takeoff and landing, and an accurate prediction of landing gear performance is beneficial to ensure flight safety. Nevertheless, the landing gear performance prediction method based on machine learning has a strong reliance on the dataset, in which the feature dimension and data distribution will have a great impact on the prediction accuracy. To address these issues, a novel MCA-MLPSA is developed. First, an MCA (multiple correlation analysis) method is proposed to select key features. Second, a heterogeneous multilearner integration framework is proposed, which makes use of different base learners. Third, an MLPSA (multilayer perceptron with self-attention) model is proposed to adaptively capture the data distribution and adjust the weights of each base learner. Finally, the excellent prediction performance of the proposed MCA-MLPSA is validated by a series of experiments on the landing gear data.

Human activity recognition (HAR) using smartphone inertial sensors, like accelerometers and gyroscopes, enhances smartphones' adaptability and user experience. Data distribution from these sensors is affected by several factors including sensor hardware, software, device placement, user demographics, terrain, and more. Most datasets focus on providing variability in user and (sometimes) device placement, limiting domain adaptation and generalization studies. Consequently, models trained on one dataset often perform poorly on others. Despite many publicly available HAR datasets, cross-dataset generalization remains challenging due to data format incompatibilities, such as differences in measurement units, sampling rates, and label encoding. Hence, we introduce the DAGHAR benchmark, a curated collection of datasets for domain adaptation and generalization studies in smartphone-based HAR. We standardized six datasets in terms of accelerometer units, sampling rate, gravity component, activity labels, user partitioning, and time window size, removing trivial biases while preserving intrinsic differences. This enables controlled evaluation of model generalization capabilities. Additionally, we provide baseline performance metrics from state-of-the-art machine learning models, crucial for comprehensive evaluations of generalization in HAR tasks. [ABSTRACT FROM AUTHOR]

The proliferation of edge devices and advancements in Internet of Things technology have created a vast array of distributed data sources, necessitating machine learning models that can operate closer to the point of data generation. Traditional centralized machine learning approaches often struggle with real-time big data applications, such as climate prediction and traffic simulation, owing to high communication costs. In this study, we introduce random node entropy pairing, a novel distributed learning method for artificial neural networks tailored to distributed computing environments. This method reduces communication overhead and addresses data imbalance by enabling nodes to exchange only the weights of their local models with randomly selected peers during each communication round, rather than sharing the entire dataset. Our findings indicate that this approach significantly reduces communication costs while maintaining accuracy, even when learning from non-IID local data. Furthermore, we explored additional learning strategies that enhance system accuracy by leveraging the characteristics of this method. The results demonstrate that random node entropy pairing is an effective and efficient solution for distributed learning in environments where communication costs and data distribution present significant challenges. [ABSTRACT FROM AUTHOR]

Financial distress prediction (FDP) has attracted high attention from many financial institutions. Utilizing supervised learning‐based methods in FDP, however, is time consuming and labor intensive. Therefore, in this paper, we exploit active‐pSVM method, which combines potential data distribution information and existing expert experience to solve FDP problem. Moreover, with the increasingly popular textual information, we construct several features on our protocol that are based on the Management Discussion and Analysis (MD&A) text information. Using datasets that are collected in different time windows from the listed Chinese companies, we conducted an extensive experiment and were able to confirm a better efficiency of our active‐pSVM, when compared with some common supervised learning‐based methods. Our study also covers the application of MD&A text information on weakly supervised learning model in FDP. [ABSTRACT FROM AUTHOR]

Background: Spatiotemporal dengue forecasting using machine learning (ML) can contribute to the development of prevention and control strategies for impending dengue outbreaks. However, training data for dengue incidence may be inflated with frequent zero values because of the rarity of cases, which lowers the prediction accuracy. This study aimed to understand the influence of spatiotemporal resolutions of data on the accuracy of dengue incidence prediction using ML models, to understand how the influence of spatiotemporal resolution differs between quantitative and qualitative predictions of dengue incidence, and to improve the accuracy of dengue incidence prediction with zero-inflated data. Methodology: We predicted dengue incidence at six spatiotemporal resolutions and compared their prediction accuracy. Six ML algorithms were compared: generalized additive models, random forests, conditional inference forest, artificial neural networks, support vector machines and regression, and extreme gradient boosting. Data from 2009 to 2012 were used for training, and data from 2013 were used for model validation with quantitative and qualitative dengue variables. To address the inaccuracy in the quantitative prediction of dengue incidence due to zero-inflated data at fine spatiotemporal scales, we developed a hybrid approach in which the second-stage quantitative prediction is performed only when/where the first-stage qualitative model predicts the occurrence of dengue cases. Principal findings: At higher resolutions, the dengue incidence data were zero-inflated, which was insufficient for quantitative pattern extraction of relationships between dengue incidence and environmental variables by ML. Qualitative models, used as binary variables, eased the effect of data distribution. Our novel hybrid approach of combining qualitative and quantitative predictions demonstrated high potential for predicting zero-inflated or rare phenomena, such as dengue. Significance: Our research contributes valuable insights to the field of spatiotemporal dengue prediction and provides an alternative solution to enhance prediction accuracy in zero-inflated data where hurdle or zero-inflated models cannot be applied. Author summary: In our study, we tackled the complex challenge of predicting dengue fever outbreaks, a crucial task in the field of epidemiology. Dengue prediction is complicated because it relies on the quality of data, which may be affected by the temporal and spatial resolution. We explored different machine learning algorithms across various spatial (village, city and region) and temporal resolutions (weekly and monthly). A key hurdle we encountered was the high frequency of zero values in reported dengue cases, a common issue known as zero-inflated data. This phenomenon makes accurate predictions difficult, especially at finer resolutions. To overcome this obstacle, we first made qualitative predictions about the presence or absence of dengue cases. Then, in scenarios indicating disease presence, we estimated the magnitude of cases quantitatively. This innovative method we designated as hybrid approach and significantly enhanced prediction accuracy in zero-inflated data. This approach can be applied to continuous data where zero-inflated or hurdle models cannot be applied. Our findings have broader implications beyond dengue prediction, shedding light on the challenges of dealing with zero-inflated data in various real-world situations. By improving our understanding of these complexities, our research contributes valuable insights that not only benefit scientists working in epidemiology but also have practical applications in public health strategies ensuring more effective and targeted interventions. [ABSTRACT FROM AUTHOR]

Introduction: The thickness of ground substrate in shallow mountainous areas is a crucial indicator for substrate investigations and a key factor in evaluating substrate quality and function. Reliable data acquisition methods are essential for effective investigation. Methods: This study utilizes six machine learning algorithms -Gradient Boosting Machine (GB), Random Forest (RF), AdaBoost Regressor (AB), Neural Network (NN), Support Vector Machine (SVM), and k-Nearest Neighbors (kNN) -to predict ground substrate thickness. Grid search optimization was employed to finetune model parameters. The models' performances were evaluated using four metrics: mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), and the coefficient of determination (R2). The optimal parameter combinations for each model were then used to calculate the spatial distribution of ground substrate thickness in the study area. Results: The results indicate that after parameter optimization, all models showed significant reductions in the MSE, RMSE, and MAE, while R2 values increased substantially. Under optimal parameters, the RF model achieved an MSE of 1,589, RMSE of 39.8, MAE of 26.5, and an R2 of 0.63, with a Pearson correlation coefficient of 0.80, outperforming the other models. Therefore, parameter tuning is a necessary step in using machine learning models to predict ground substrate thickness, and the performance of all six models improved significantly after tuning. Overall, ensemble learning models provided better predictive performance than other machine learning models, with the RF model demonstrating the best accuracy and robustness. Discussion: Moreover, further attention is required on the characteristics of sample data and environmental variables in machine learning-based predictions. [ABSTRACT FROM AUTHOR]

The article discusses disruptions in clinical trials caused by external events like pandemics, warfare, and natural disasters, leading to complications such as site closures, supply chain interruptions, and travel restrictions. It explores the use of randomization tests to address disruptions in clinical trials and highlights the importance of incorporating trial disruptions as intercurrent events in the statistical analysis plan. The article also introduces Targeted Learning (TL) as a method to handle disruptions in clinical trials, emphasizing the need for robust analytic strategies in the statistical analysis plan. TL provides a guide for statistical and causal inference, aligning with the principles set forth in the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) E9(R1) addendum. [Extracted from the article]

This paper proposes a novel method capable of both detecting OOD data and generating in-distribution data samples. To achieve this, a VAE model is adopted and augmented with a memory module, providing capacities for identifying OOD data and synthesising new in-distribution samples. The proposed VAE is trained on normal data and the memory stores prototypical patterns of the normal data distribution. At test time, the input is encoded by the VAE encoder; this encoding is used as a query to retrieve related memory items, which are then integrated with the input encoding and passed to the decoder for reconstruction. Normal samples reconstruct well and yield low reconstruction errors, while OOD inputs produce high reconstruction errors as their encodings get replaced by retrieved normal patterns. Prior works use memory modules for OOD detection with autoencoders, but this method leverages a VAE architecture to enable generation abilities. Experiments conducted with CIFAR-10 and MNIST datasets show that the memory-augmented VAE consistently outperforms the baseline, particularly where OOD data resembles normal patterns. This notable improvement is due to the enhanced latent space representation provided by the VAE. Overall, the memory-equipped VAE framework excels in identifying OOD and generating creative examples effectively. [ABSTRACT FROM AUTHOR]

Imbalanced datasets often lead to biased machine learning models that underperform on minority classes. The Synthetic Minority Over-sampling Technique (SMOTE) addresses this by generating synthetic samples for the minority class. Despite its empirical success, the theoretical properties of SMOTE remain underexplored. This paper investigates the asymptotic behavior of SMOTE-generated random variables using order statistics. We establish that, as the sample size n increases, the SMOTE-generated variable Z conditioned on the k-th order statistic X(k) converges in mean to X(k). Additionally, the expected value of Z converges to the expected value of the original random variable X as n approaches infinity. The results are derived under the assumption of left-bounded support for X. Our findings provide a theoretical foundation for SMOTE, demonstrating that it preserves key statistical properties of the original data distribution in large samples. [ABSTRACT FROM AUTHOR]

In this paper, we propose systematic approaches for learning imbalanced data based on a two-regime process: regime 0, which generates excess zeros (majority class), and regime 1, which contributes to generating an outcome of one (minority class). The proposed model contains two latent equations: a split probit (logit) equation in the first stage and an ordinary probit (logit) equation in the second stage. Because boosting improves the accuracy of prediction versus using a single classifier, we combined a boosting strategy with the two-regime process. Thus, we developed the zero-inflated probit boost (ZIPBoost) and zero-inflated logit boost (ZILBoost) methods. We show that the weight functions of ZIPBoost have the desired properties for good predictive performance. Like AdaBoost, the weight functions upweight misclassified examples and downweight correctly classified examples. We show that the weight functions of ZILBoost have similar properties to those of LogitBoost. The algorithm will focus more on examples that are hard to classify in the next iteration, resulting in improved prediction accuracy. We provide the relative performance of ZIPBoost and ZILBoost, which rely on the excess kurtosis of the data distribution. Furthermore, we show the convergence and time complexity of our proposed methods. We demonstrate the performance of our proposed methods using a Monte Carlo simulation, mergers and acquisitions (M&A) data application, and imbalanced datasets from the Keel repository. The results of the experiments show that our proposed methods yield better prediction accuracy compared to other learning algorithms. [ABSTRACT FROM AUTHOR]

Network anomaly detection plays a vital role in safeguarding network security. However, the existing network anomaly detection task is typically based on the one-class zero-positive scenario. This approach is susceptible to overfitting during the training process due to discrepancies in data distribution between the training set and the test set. This phenomenon is known as prediction drift. Additionally, the rarity of anomaly data, often masked by normal data, further complicates network anomaly detection. To address these challenges, we propose the PUNet network, which ingeniously combines the strengths of traditional machine learning and deep learning techniques for anomaly detection. Specifically, PUNet employs a reconstruction-based autoencoder to pre-train normal data, enabling the network to capture potential features and correlations within the data. Subsequently, PUNet integrates a sampling algorithm to construct a pseudo-label candidate set among the outliers based on the reconstruction loss of the samples. This approach effectively mitigates the prediction drift problem by incorporating abnormal samples. Furthermore, PUNet utilizes the CatBoost classifier for anomaly detection to tackle potential data imbalance issues within the candidate set. Extensive experimental evaluations demonstrate that PUNet effectively resolves the prediction drift and data imbalance problems, significantly outperforming competing methods. [ABSTRACT FROM AUTHOR]

Generative design refers to a methodology that not only simulates the characteristics of a given data or system but also creates artificial data for various purposes. It's a significant research area encompassing diverse issues such as privacy preservation, data distribution analysis, and the development of surrogate models. Initially, research in this field primarily employed stochastic models or basic machine learning methods. However, with the advancement of deep learning technology, numerous studies have emerged, showcasing developed mechanisms using artificial neural network-based methods like variational autoencoders (VAEs) and generative adversarial networks (GANs). These studies extend across different data types, including images and texts, tailored to specific objectives. This paper presents a systematic review of generative design research focused on tabular data. We begin by elucidating the characteristics of tabular data within generative design, followed by a discussion on the goals and challenges in this area. Subsequently, the paper introduces various generative design studies on tabular data, categorized according to their methodological development and unique objectives. Finally, we address the benchmark methods used in generative design for tabular and how their performance is evaluated. [ABSTRACT FROM AUTHOR]

Effective soil conditioning is critical for mechanized shield tunneling, yet the selection of conditioning parameters remains experience‐oriented. This study presents a machine learning–informed soil conditioning strategy, aiming at enabling automatic decision‐making for soil conditioning during tunneling. The proposed procedure includes feature engineering to process raw data, the selection of an optimal model, and a committee‐based uncertainty quantification strategy to evaluate the reliability of models. High‐dimensional data visualization techniques are leveraged to examine the influence of data distribution on the model's performance. Results demonstrate that feature engineering and the selection of models lay a foundation for machine learning–enabled automatic soil conditioning, while historical soil conditioning parameters should be incorporated in models to account for time‐dependent features during tunneling. The committee‐based uncertainty quantification strategy can effectively identify the reliability of both interpolated and extrapolated predictions. The proposed data‐driven soil conditioning framework can promote the application of machine learning toward automating earth pressure balance (EPB) shield tunneling. [ABSTRACT FROM AUTHOR]

We study the interplay between the data distribution and Q-learning-based algorithms with function approximation. We provide a unified theoretical and empirical analysis as to how different properties of the data distribution influence the performance of Q-learning-based algorithms. We connect different lines of research, as well as validate and extend previous results, being primarily focused on offline settings. First, we analyze the impact of the data distribution by using optimization as a tool to better understand which data distributions yield low concentrability coefficients. We motivate high-entropy distributions from a game-theoretical point of view and propose an algorithm to find the optimal data distribution from the point of view of concentrability. Second, from an empirical perspective, we introduce a novel four-state MDP specifically tailored to highlight the impact of the data distribution in the performance of Q-learning-based algorithms with function approximation. Finally, we experimentally assess the impact of the data distribution properties on the performance of two offline Q-learning-based algorithms under different environments. Our results attest to the importance of different properties of the data distribution such as entropy, coverage, and data quality (closeness to optimal policy). [ABSTRACT FROM AUTHOR]

This study investigates the feasibility of using a basic 24 GHz one-dimensional (1D) radar for Human Activity Recognition (HAR), focusing on differentiating between walking and standing movements. We evaluate the radar's performance using various machine learning models, including K-means, GMM, SVM, and LSTM. Using the silhouette score and the Davies-Bouldin index, we evaluate the intra- and interclass results of K-means and GMM, while SVM and LSTM are used to analyze their performance. The results indicate that the LSTM model achieves high accuracy in both vertical and horizontal dimensions, with precision, recall, and F1-scores all above 98% for both standing and walking movements. However, the SVM model faces challenges in horizontal movement detection, consistent with the unsupervised learning results where the inter-class and intra-class distances for the horizontal dimension are not significant, making differentiation difficult. These findings delineate the boundaries and capabilities of a lower-specification radar for HAR, providing insights into its practical applications and limitations. [ABSTRACT FROM AUTHOR]

With the emergence and development of application requirements such as data analysis and publishing, it is particularly important to use differential privacy protection technology to provide more reliable, secure, and compliant datasets for research in the field of children's health. This paper focuses on the differential privacy protection of the ultrasound examination health dataset of adolescents in southern Texas from three aspects: differential privacy protection with output perturbation on basic statistics, publication of differential privacy marginal histogram and synthesized data, and a machine learning differential privacy learning algorithm. Firstly, differential privacy protection results with output perturbation show that Laplace and Gaussian mechanisms for numerical data, as well as the exponential mechanism for non-numerical data, can achieve the goal of protecting privacy. The exponential mechanism provides higher privacy protection. Secondly, a differential privacy marginal histogram with four attributes can be obtained with an appropriate privacy budget that approximates the marginal histogram of the original data. In order to publish synthetic data, we construct a synthetic query to obtain the corresponding differential privacy histogram for two attributes. Further, a synthetic dataset can be constructed by following the data distribution of the original dataset and the quality of the synthetic data publication can also be evaluated by the mean square error and error rate. Finally, consider a differential privacy logistic regression model under machine learning to predict whether children have fatty liver in binary classification tasks. The experimental results show that the model combined with quadratic perturbation has better accuracy and privacy protection. This paper can provide differential privacy protection models under different demands, which provides important data release and analysis options for data managers and research organizations, in addition to enriching the research on child health data releasing and mining. [ABSTRACT FROM AUTHOR]

Background: Diabetes, as a significant disease affecting public health, requires early detection for effective management and intervention. However, imbalanced datasets pose a challenge to accurate diabetes prediction. This imbalance often results in models performing poorly in predicting minority classes, affecting overall diagnostic performance. Objectives: To address this issue, this study employs a combination of Synthetic Minority Over-sampling Technique (SMOTE) and Random Under-Sampling (RUS) for data balancing and uses Optuna for hyperparameter optimization of machine learning models. This approach aims to fill the gap in current research concerning data balancing and model optimization, thereby improving prediction accuracy and computational efficiency. Methods: First, the study uses SMOTE and RUS methods to process the imbalanced diabetes dataset, balancing the data distribution. Then, Optuna is utilized to optimize the hyperparameters of the LightGBM model to enhance its performance. During the experiment, the effectiveness of the proposed methods is evaluated by comparing the training results of the dataset before and after balancing. Results: The experimental results show that the enhanced LightGBM-Optuna model improves the accuracy from 97.07% to 97.11%, and the precision from 97.17% to 98.99%. The time required for a single search is only 2.5 seconds. These results demonstrate the superiority of the proposed method in handling imbalanced datasets and optimizing model performance. Conclusions: The study indicates that combining SMOTE and RUS data balancing algorithms with Optuna for hyperparameter optimization can effectively enhance machine learning models, especially in dealing with imbalanced datasets for diabetes prediction. [ABSTRACT FROM AUTHOR]

Distributed machine learning has received much attention for more than two decades. Yet, it is still a challenge to achieve acceptable performance in practical scenarios when some features of data samples are missing. Although some imputation methods have been proposed for handling missing data, their performance deteriorates significantly when data are heterogeneously distributed over different nodes in the network. Considering this, in this article, we first propose a general Gaussian mixture model (GMM) consisting of both public and personalized components for modeling the homogeneous and the heterogeneous parts of data distribution, respectively. Then, we develop a distributed personalized expectation–maximization method based on knowledge transfer (KT-dpEM) to estimate the parameters of the proposed general GMM. After that, based on the estimated general GMM, missing data are imputed using the posterior conditional mean. Experimental results show that the proposed KT-dpEM algorithm has better imputation accuracy, higher robustness against different missing probabilities and better classification performance in the downstream classification tasks, compared with state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Imbalanced data poses a significant challenge in machine learning, as conventional classification algorithms often prioritize majority class samples, while accurately classifying minority class samples is more crucial. The synthetic minority oversampling technique (SMOTE) represents one of the most renowned methods for handling imbalanced data. However, both SMOTE and its variants have limitations due to their insufficient consideration of data distribution, leading to the generation of incorrect and unnecessary samples. This paper, therefore, introduces a novel oversampling algorithm called data distribution and spectral clustering-based SMOTE (DDSC-SMOTE). This algorithm addresses the shortcomings of SMOTE by introducing three innovative data distribution-based improvement strategies: adaptive allocation of synthetic sample quantities strategy, seed sample adaptive selection strategy, and synthetic sample improvement strategy. First, we use the k-nearest neighbor sample labels and the local outlier factor algorithm to remove noisy and outlier samples. Next, we leverage spectral clustering to identify clusters within the minority class and propose a dual-weight factor that considers inter-cluster and intra-cluster distances to allocate the number of synthetic samples effectively, addressing interclass and intraclass imbalances. Furthermore, we introduce a relative position weight coefficient to determine the probability of selecting seed samples within the subcluster, ensuring that important minority samples have higher chances of being sampled. Finally, we improve the SMOTE sample synthesis formula for safer generation. Extensive comparisons on real datasets from the UCI repository demonstrate that DDSC-SMOTE outperforms seven state-of-the-art oversampling algorithms significantly in terms of G-mean and F1-score, presenting a data distribution-focused solution for addressing imbalanced data challenges. [ABSTRACT FROM AUTHOR]

This paper proposes a learning-based model for the resource-constrained edge nodes in the blockchain-enabled Internet of Medical Things (IoMT) systems to realize efficient querying. Three layers are designed in the new model: data evaluation layer, data storage layer and data distribution layer. The data evaluation layer extracts the features from medical data and evaluates their values based on the Extreme Learning Machine (ELM) method. Then, in the data storage layer, according to the value of medical data, a novelty data structure called Merkle–Huffman tree (M-H tree) is established. Compared with the Merkle tree, high-value data (frequently accessed data) in M-H tree is saved closer to the root node and can be found faster. In the data distribution layer, the sharding-based blockchain model is adopted to increase the storage scalability of the IoMT system. Finally, the experimental results show that the new learning-based model can effectively improve the query speed of the blockchain-enabled medical system by about 3.5% and free up large amounts of storage space on IoMT devices. [ABSTRACT FROM AUTHOR]

Recently, advanced AI systems equipped with sophisticated learning algorithms have emerged, enabling the processing of extensive streaming data for online decision-making in diverse domains. However, the widespread deployment of these systems has prompted concerns regarding potential ethical issues, particularly the risk of discrimination that can adversely impact certain community groups. This issue has been proven to be challenging to address in the context of streaming data, where data distribution can change over time, including changes in the level of discrimination within the data. In addition, transparent models like decision trees are favoured in such applications because they illustrate the decision-making process. However, it is essential to keep the models compact because the explainability of large models can diminish. Existing methods usually mitigate discrimination at the cost of accuracy. Accuracy and discrimination, therefore, can be considered conflicting objectives. Current methods are still limited in controlling the trade-off between these conflicting objectives. This paper proposes a method that can incrementally learn classification models from streaming data and automatically adjust the learnt models to balance multi-objectives simultaneously. The novelty of this research is to propose a multi-objective algorithm to maximise accuracy, minimise discrimination and model size simultaneously based on swarm intelligence. Experimental results using six real-world datasets show that the proposed algorithm can evolve fairer and simpler classifiers while maintaining competitive accuracy compared to existing state-of-the-art methods tailored for streaming data. [ABSTRACT FROM AUTHOR]

In order to solve the problems of low and unstable prediction accuracy of the current yarn quality prediction model based on neural network, three prediction models were established, including RF (random forest) algorithm prediction model, MLP (multi-layer perceptron) neural network algorithm prediction model and LR (linear regression) algorithm prediction model. Experiments were conducted to compare the three models' prediction performance on the sensitivity with small samples in different data distribution, the sensitivity in different data dimensions and the sensitivity in different training sample sizes. The coefficient of determination and root mean square error were used to evaluate the prediction performance of each model. The experimental results showed that, compared with MLP algorithm and LR algorithm, in most cases, RF algorithm had higher prediction accuracy, better stability of prediction accuracy, better adaptability to small training sample size, and higher comprehensive prediction performance. [ABSTRACT FROM AUTHOR]

The accurate prediction of global horizontal irradiance (GHI) is crucial for optimizing solar power generation systems, particularly in mountainous areas with complex topography and unique microclimates. These regions face significant challenges due to limited reliable data and the dynamic nature of local weather conditions, which complicate accurate GHI measurement. The scarcity of precise data impedes the development of reliable solar energy prediction models, impacting both economic and environmental outcomes. To address these data scarcity challenges in solar energy prediction, this paper focuses on various locations in Europe and Asia Minor, predominantly in mountainous regions. Advanced machine learning techniques, including random forest (RF) and extreme gradient boosting (XGBoost) regressors, are employed to effectively predict GHI. Additionally, optimizing training data distribution based on cloud opacity values and integrating synthetic data significantly enhance predictive accuracy, with R2 scores ranging from 0.91 to 0.97 across multiple locations. Furthermore, substantial reductions in root mean square error (RMSE), mean absolute error (MAE), and mean bias error (MBE) underscore the improved reliability of the predictions. Future research should refine synthetic data generation, optimize additional meteorological and environmental parameter integration, extend methodology to new regions, and test for predicting global tilted irradiance (GTI). The studies should expand training data considerations beyond cloud opacity, incorporating sky cover and sunshine duration to enhance prediction accuracy and reliability. [ABSTRACT FROM AUTHOR]

Machine learning is increasingly applied to Earth Observation (EO) data to obtain datasets that contribute towards international accords. However, these datasets contain inherent uncertainty that needs to be quantified reliably to avoid negative consequences. In response to the increased need to report uncertainty, we bring attention to the promise of conformal prediction within the domain of EO. Unlike previous uncertainty quantification methods, conformal prediction offers statistically valid prediction regions while concurrently supporting any machine learning model and data distribution. To support the need for conformal prediction, we reviewed EO datasets and found that only 22.5% of the datasets incorporated a degree of uncertainty information, with unreliable methods prevalent. Current open implementations require moving large amounts of EO data to the algorithms. We introduced Google Earth Engine native modules that bring conformal prediction to the data and compute, facilitating the integration of uncertainty quantification into existing traditional and deep learning modelling workflows. To demonstrate the versatility and scalability of these tools we apply them to valued EO applications spanning local to global extents, regression, and classification tasks. Subsequently, we discuss the opportunities arising from the use of conformal prediction in EO. We anticipate that accessible and easy-to-use tools, such as those provided here, will drive wider adoption of rigorous uncertainty quantification in EO, thereby enhancing the reliability of downstream uses such as operational monitoring and decision-making. [ABSTRACT FROM AUTHOR]

Distribution shift, a phenomenon in machine learning characterized by a change in input data distribution between training and testing, can reduce the predictive accuracy of deep learning models. As operator and hardware conditions at the time of training are not always consistent with those after deployment, computer vision wood identification (CVWID) models are potentially susceptible to the negative impacts of distribution shift in the field. To maximize the robustness of CVWID models, it is critical to evaluate the influence of distribution shifts on model performance. In this study, a previously published 24-class CVWID model for Peruvian timbers was evaluated on images of test specimens digitally perturbed to simulate four kinds of image variations an operator might encounter in the field including (1) red and blue color shifts to simulate sensor drift or the effects of disparate sensors; (2) resizing to simulate different magnifications that could result from using different or improperly calibrated hardware; (3) digital scratches to simulate artifacts of specimen preparation; and (4) a range of blurring effects to simulate out-of-focus images. The model was most robust to digital scratches, moderately robust to red shift and smaller areas of medium-to-severe blur, and was least robust to resizing, blue shift, and large areas of medium-to-severe blur. These findings emphasize the importance of formulating and consistently applying best practices to reduce the occurrence of distribution shift in practice and standardizing imaging hardware and protocols to ensure dataset compatibility across CVWID platforms. [ABSTRACT FROM AUTHOR]

As a significant application of machine learning in financial scenarios, loan default risk prediction aims to evaluate the client's default probability. However, most existing deep learning solutions treat each application as an independent individual, neglecting the explicit connections among different application records. Besides, these attempts suffer from the problem of missing data and imbalanced distribution (i.e., the default records are small samples against all the applications). We believe similar records could provide some auxiliary signals, which are of critical importance to alleviate the data missing issue and facilitate data argumentation. To this end, we propose multi-view loan application graphs, dubbed MLAGs. By evaluating the similarity between the records, a loan application graph can be constructed. Furthermore, we arrange different similarity thresholds to organize various graph structures for multi-graph constructions; thus, a variety of representations can be generated via information propagation and aggregation for small sample argumentation. Consequently, the imbalanced data distribution and missing values issues can be alleviated effectively. We conduct experiments on three public datasets from real-world home credit and P2P lending platforms, which show that MGCN outperforms both conventional and deep learning models. Ablation studies also illustrated the validity of each module design. [ABSTRACT FROM AUTHOR]

Traditional machine learning techniques depend on both the training data and the target data having the same data distribution and feature space. The performance may be unsatisfactory if there is a discrepancy in data distribution between the training data and the target data, which is known as a cross-domain learning issue. In this study, a technique for addressing this issue based on using the envelope analysis features coded as inputs to the Artificial Neural Networks (ANN) is presented. Specifically, three major cross-domain difficulties were resolved, including defect identification of different roller bearing types, rotating speeds, and loading conditions. The envelope approach is used to extract more distinguishing features at fundamental fault frequencies from the original signal, with the benefit of derived features being independent of both roller bearing type and operating conditions of rotating machines. The diagnosis outcomes were achieved experimentally through three different types of rollers bearing. Meanwhile, for each bearing type, three data sets were obtained at three different rotational speeds with different levels of fault categories to simulate all the cross-domain tasks. Moreover, the ANN model was trained using data of a particular bearing type and a certain operating state, whereas data of other bearing types were utilized to test the cross-domain performance of the proposed technique. At the same loading condition, the results indicate a 99.5% average success rate for bearing Koyo 1205, a 98.33% average accuracy for bearing NU 205, and a 97.3% average accuracy for the defective bearing kit. Furthermore, the results indicate that the suggested technique can deliver accurate cross-domain detection. Also, the experimental results proved that the suggested technology is a potential strategy for industrial applications. [ABSTRACT FROM AUTHOR]

Recent studies have extensively used deep learning algorithms to analyze gene expression to predict disease diagnosis, treatment effectiveness, and survival outcomes. Survival analysis studies on diseases with high mortality rates, such as cancer, are indispensable. However, deep learning models are plagued by overfitting owing to the limited sample size relative to the large number of genes. Consequently, the latest style-transfer deep generative models have been implemented to generate gene expression data. However, these models are limited in their applicability for clinical purposes because they generate only transcriptomic data. Therefore, this study proposes ctGAN, which enables the combined transformation of gene expression and survival data using a generative adversarial network (GAN). ctGAN improves survival analysis by augmenting data through style transformations between breast cancer and 11 other cancer types. We evaluated the concordance index (C-index) enhancements compared with previous models to demonstrate its superiority. Performance improvements were observed in nine of the 11 cancer types. Moreover, ctGAN outperformed previous models in seven out of the 11 cancer types, with colon adenocarcinoma (COAD) exhibiting the most significant improvement (median C-index increase of ~15.70%). Furthermore, integrating the generated COAD enhanced the log-rank p -value (0.041) compared with using only the real COAD (p -value = 0.797). Based on the data distribution, we demonstrated that the model generated highly plausible data. In clustering evaluation, ctGAN exhibited the highest performance in most cases (89.62%). These findings suggest that ctGAN can be meaningfully utilized to predict disease progression and select personalized treatments in the medical field. [ABSTRACT FROM AUTHOR]

Class imbalance has been widely accepted as a significant factor that negatively impacts a machine learning classifier's performance. One of the techniques to avoid this problem is to balance the data distribution by using sampling-based approaches, in which synthetic data is generated using the probability distribution of the classes. However, this process is sensitive to the presence of noise in the data, and the boundaries between the majority class and the minority class are blurred. Such phenomena shift the algorithm's decision boundary away from the ideal outcome. In this work, we propose a hybrid framework for two primary objectives. The first objective is to address class distribution imbalance by synthetically increasing the data of a minority class, and the second objective is, to devise an efficient noise reduction technique that improves the class balance algorithm. The proposed framework focuses on removing noisy elements from the majority class, and by doing so, provides more accurate information to the subsequent synthetic data generator algorithm. To evaluate the effectiveness of our framework, we employ the geometric mean (G-mean) as the evaluation metric. The experimental results show that our framework is capable of improving the prediction G-mean for eight classifiers across eleven datasets. The range of improvements varies from 7.78% on the Loan dataset to 67.45% on the Abalone19_vs_10-11-12-13 dataset. [ABSTRACT FROM AUTHOR]

Modeling housing markets is a challenging and central research area since they are highly related to the economy. However, the limited available data prevents researchers from improving models. As an alternative, this study introduces Housing GANs, a data-driven modeling approach inspired by the recent success of generative adversarial networks (GANs). The Housing GANs include a generator and discriminator function utilizing Wasserstein GAN with gradient penalty and mitigate original housing datasets, including continuous and discrete data. The generator function predicts the real data distribution and generates realistic housing data. The empirical analysis highlights that the Housing GANs successfully learns the distribution and generate realistic housing data in high fidelity. [ABSTRACT FROM AUTHOR]

As the user's behavior changes at any time with cloud computing and network services, abnormal server resource utilization traffic will lead to severe service crashes and system downtime. The traditional single anomaly detection model cannot handle the rapid failure prediction ahead. Therefore, this study proposed ensemble learning combined with model-agnostic meta-reinforcement learning called ensemble meta-reinforcement learning (EMRL) to implement self-adaptive server anomaly detection rapidly and precisely, according to the time series of server resource utilization. The proposed ensemble approach combines hidden Markov model (HMM), variational autoencoder (VAE), temporal convolutional autoencoder (TCN-AE), and bidirectional long short-term memory (BLSTM). The EMRL algorithm trains this combination with several tasks to learn the implicit representation of various anomalous traffic, where each task executes trust region policy optimization (TRPO) to quickly adapt the time-varying data distribution and make rapid decisions precisely for an agent response. As a result, our proposed approach can improve the precision of anomaly prediction by 2.4 times and reduce the model deployment speed by 5.8 times on average because a meta-learner can immediately be applied to new tasks. [ABSTRACT FROM AUTHOR]

In contemporary society, depression has emerged as a prominent mental disorder that exhibits exponential growth and exerts a substantial influence on premature mortality. Although numerous research applied machine learning methods to forecast signs of depression. Nevertheless, only a limited number of research have taken into account the severity level as a multiclass variable. Besides, maintaining the equality of data distribution among all the classes rarely happens in practical communities. So, the inevitable class imbalance for multiple variables is considered a substantial challenge in this domain. Furthermore, this research emphasizes the significance of addressing class imbalance issues in the context of multiple classes. We introduced a new approach Feature group partitioning (FGP) in the data preprocessing phase which effectively reduces the dimensionality of features to a minimum. This study utilized synthetic oversampling techniques, specifically Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic (ADASYN), for class balancing. The dataset used in this research was collected from university students by administering the Burn Depression Checklist (BDC). For methodological modifications, we implemented heterogeneous ensemble learning stacking, homogeneous ensemble bagging, and five distinct supervised machine learning algorithms. The issue of overfitting was mitigated by evaluating the accuracy of the training, validation, and testing datasets. To justify the effectiveness of the prediction models, balanced accuracy, sensitivity, specificity, precision, and f1-score indices are used. Overall, comprehensive analysis demonstrates the discrimination between the Conventional Depression Screening (CDS) and FGP approach. In summary, the results show that the stacking classifier for FGP with SMOTE approach yields the highest balanced accuracy, with a rate of 92.81%. The empirical evidence has demonstrated that the FGP approach, when combined with the SMOTE, able to produce better performance in predicting the severity of depression. Most importantly the optimization of the training time of the FGP approach for all of the classifiers is a significant achievement of this research. [ABSTRACT FROM AUTHOR]

Due to its plug-and-play functionality and wide device support, the universal serial bus (USB) protocol has become one of the most widely used protocols. However, this widespread adoption has introduced a significant security concern: the implicit trust provided to USB devices, which has created a vast array of attack vectors. Malicious USB devices exploit this trust by disguising themselves as benign peripherals and covertly implanting malicious commands into connected host devices. Existing research employs supervised learning models to identify such malicious devices, but our study reveals a weakness in these models when faced with sophisticated data poisoning attacks. We propose, design and implement a sophisticated adversarial data poisoning attack to demonstrate how these models can be manipulated to misclassify an attack device as a benign device. Our method entails generating keystroke data using a microprogrammable keystroke attack device. We develop adversarial attacker by meticulously analyzing the data distribution of data features generated via USB keyboards from benign users. The initial training data is modified by exploiting firmware-level modifications within the attack device. Upon evaluating the models, our findings reveal a significant decrease from 99 to 53% in detection accuracy when an adversarial attacker is employed. This work highlights the critical need to reevaluate the dependability of machine learning-based USB threat detection mechanisms in the face of increasingly sophisticated attack methods. The vulnerabilities demonstrated highlight the importance of developing more robust and resilient detection strategies to protect against the evolution of malicious USB devices. [ABSTRACT FROM AUTHOR]

The quality of machine learning (ML) models deployed in dynamic environments tends to decline over time due to disparities between the data used for training and the upcoming data available for prediction, which is commonly known as drift. Therefore, it is important for ML models to be capable of detecting any changes or drift in the data distribution and updating the ML model accordingly. This study presents various drift detection techniques to identify drift in the survival outcomes of patients who underwent cardiac surgery. Additionally, this study proposes several drift adaptation strategies, such as adaptive learning, incremental learning, and ensemble learning. Through a detailed analysis of the results, the study confirms the superior performance of ensemble model, achieving a minimum mean absolute error (MAE) of 10.684 and 2.827 for predicting hospital stay and ICU stay, respectively. Furthermore, the models that incorporate a drift adaptive framework exhibit superior performance compared to the models that do not include such a framework. [ABSTRACT FROM AUTHOR]

This study introduces a novel algorithm tailored for the precise detection of lower outliers (i.e., data points at the lower tail) in univariate datasets, which is particularly suited for scenarios with a single cluster and similar data distribution. The approach leverages a combination of transformative techniques and advanced filtration methods to efficiently segregate anomalies from normal values. Notably, the algorithm emphasizes high-precision outlier detection, ensuring minimal false positives, and requires only a few parameters for configuration. Its unsupervised nature enables robust outlier filtering without the need for extensive manual intervention. To validate its efficacy, the algorithm is rigorously tested using real-world data obtained from photovoltaic (PV) module strings with similar DC capacities, containing various outliers. The results demonstrate the algorithm's capability to accurately identify lower outliers while maintaining computational efficiency and reliability in practical applications. [ABSTRACT FROM AUTHOR]

Semi-supervised learning has achieved extraordinary success in prevalent image-classification benchmarks. However, a class-balanced distribution that differs notably from real-world data distribution is required. In general, models trained under class-imbalanced semi-supervised learning conditions are severely biased towards the majority classes. To address this issue, we propose a novel framework called ABAE by implanting an Auxiliary Balanced AutoEncoder branch into existing semi-supervised learning algorithms. Considering that adaptive feature augmentation for different classes can alleviate confirmation bias, we devise a class-aware reconstruction loss to train the AutoEncoder module. To further smooth the output, we adopt a graph-based label propagation scheme at the end of the AutoEncoder. Extensive experiments on CIFAR-10/100-LT, SVHN-LT and Small ImageNet-127 demonstrate the effectiveness of ABAE. • Propose an Auxiliary Balanced AutoEncoder branch(ABAE) with a class-aware reconstruction loss for CISSL. • Utilize the adaptive feature augmentation for different classes to alleviate the confirmation bias. • Adopt a graph-based label propagation scheme to smooth the output of the ABAE. • Demonstrate the effectiveness of the ABAE through experiments on various imbalanced datasets. [ABSTRACT FROM AUTHOR]

We study learning from unlabeled data distributed across clients in a federated fashion where raw data do not leave the corresponding devices. We develop a K-means clustering algorithm within this federated setting where the local datasets are clustered at the clients, and a server generates the global clusters after aggregating the local ones. Given the importance of initialization on the federated K ̲ -means algorithm (FKM), our objective is to find better initial centroids by utilizing the local data stored on each client. To this end, we start the centroid initialization at the clients, rather than at the server, since the server initially lacks any preliminary insight into the clients' data. The clients first select their local initial clusters and subsequently share their clustering information (including cluster centroids and sizes) with the server. The server then employs a greedy algorithm to determine the global initial centroids based on the information received from the clients. We refer to this idea as G-FKM. Numerical results obtained from both synthetic and public datasets demonstrate that our proposed algorithm demonstrates accelerated convergence, exhibiting reduced within-cluster sum of squares ( WCSS ¯ ) and higher adjusted Rand Index compared to three distinct federated K-means variants. This improvement comes at a relatively low cost of sending limited additional information from the clients to the server, rather than conducting the initialization entirely at the server. Furthermore, we have also observed that the proposed algorithm performs better than the centralized algorithm for cases where the data distribution across the clients is highly skewed. [ABSTRACT FROM AUTHOR]

Federated learning (FL) is a promising approach for learning a model from data distributed on massive clients without exposing data privacy. It works effectively in the ideal federation where clients share homogeneous data distribution and learning behavior. However, FL may fail to function appropriately when the federation is not ideal, amid an unhealthy state called Negative Federated Learning (NFL), in which most clients gain no benefit from participating in FL. Many studies have tried to address NFL. However, their solutions either (1) predetermine to prevent NFL in the entire learning life-cycle or (2) tackle NFL in the aftermath of numerous learning rounds. Thus, they either (1) indiscriminately incur extra costs even if FL can perform well without such costs or (2) waste numerous learning rounds. Additionally, none of the previous work takes into account the clients who may be unwilling/unable to follow the proposed NFL solutions when using those solutions to upgrade an FL system in use. This paper introduces FL-GUARD, a holistic framework that can be employed on any FL system for tackling NFL in a run-time paradigm. That is, to dynamically detect NFL at the early stage (tens of rounds) of learning and then to activate recovery measures when necessary. Specifically, we devise a cost-effective NFL detection mechanism, which relies on an estimation of performance gain on clients. Only when NFL is detected, we activate the NFL recovery process, in which each client learns in parallel an adapted model when training the global model. Extensive experiment results confirm the effectiveness of FL-GUARD in detecting NFL and recovering from NFL to a healthy learning state. We also show that FL-GUARD is compatible with previous NFL solutions and robust against clients unwilling/unable to take any recovery measures. [ABSTRACT FROM AUTHOR]

This research study aims to understand the application of Artificial Neural Networks (ANNs) to forecast the Self-Compacting Recycled Coarse Aggregate Concrete (SCRCAC) compressive strength. From different literature, 602 available data sets from SCRCAC mix designs are collected, and the data are rearranged, reconstructed, trained and tested for the ANN model development. The models were established using seven input variables: the mass of cementitious content, water, natural coarse aggregate content, natural fine aggregate content, recycled coarse aggregate content, chemical admixture and mineral admixture used in the SCRCAC mix designs. Two normalization techniques are used for data normalization to visualize the data distribution. For each normalization technique, three transfer functions are used for modelling. In total, six different types of models were run in MATLAB and used to estimate the 28th day SCRCAC compressive strength. Normalization technique 2 performs better than 1 and TANSING is the best transfer function. The best k-fold cross-validation fold is k = 7. The coefficient of determination for predicted and actual compressive strength is 0.78 for training and 0.86 for testing. The impact of the number of neurons and layers on the model was performed. Inputs from standards are used to forecast the 28th day compressive strength. Apart from ANN, Machine Learning (ML) techniques like random forest, extra trees, extreme boosting and light gradient boosting techniques are adopted to predict the 28th day compressive strength of SCRCAC. Compared to ML, ANN prediction shows better results in terms of sensitive analysis. The study also extended to determine 28th day compressive strength from experimental work and compared it with 28th day compressive strength from ANN best model. Standard and ANN mix designs have similar fresh and hardened properties. The average compressive strength from ANN model and experimental results are 39.067 and 38.36 MPa, respectively with correlation coefficient is 1. It appears that ANN can validly predict the compressive strength of concrete. [ABSTRACT FROM AUTHOR]

Passive acoustic monitoring (PAM) studies generate thousands of hours of audio, which may be used to monitor specific animal populations, conduct broad biodiversity surveys, detect threats such as poachers, and more. Machine learning classifiers for species identification are increasingly being used to process the vast amount of audio generated by bioacoustic surveys, expediting analysis and increasing the utility of PAM as a management tool. In common practice, a threshold is applied to classifier output scores, and scores above the threshold are aggregated into a detection count. The choice of threshold produces biased counts of vocalizations, which are subject to false positive/ negative rates that may vary across subsets of the dataset. In this work, we advocate for directly estimating call density: The proportion of detection windows containing the target vocalization, regardless of classifier score. We propose a validation scheme for estimating call density in a body of data and obtain, through Bayesian reasoning, probability distributions of confidence scores for both the positive and negative classes. We use these distributions to predict site-level call densities, which may be subject to distribution shifts (when the defining characteristics of the data distribution change). These methods may be applied to the outputs of any binary detection classifier operating on fixedsize audio input windows. We test our proposed methods on a real-world study of Hawaiian birds and provide simulation results leveraging existing fully annotated datasets, demonstrating robustness to variations in call density and classifier model quality. [ABSTRACT FROM AUTHOR]

Imbalanced data distribution and label correlation are two intrinsic characteristics of multi-label data. This occurs because in this type of data, instances associated with certain labels may be sparse, and some labels may be associated with others, posing a challenge for traditional machine learning techniques. To simultaneously adapt imbalanced data distribution and label correlation, this study proposed a novel algorithm called compensation-based correlated k -labelsets (CC k EL). First, for each label, the CC k EL selects the k -1 strongest correlated labels in the label space to constitute multiple correlated k -labelsets; this improves its efficiency in comparison with the random k -labelsets (RA k EL) algorithm. Then, the CC k EL transforms each k -labelset into a multiclass issue. Finally, it uses a fast decision output compensation strategy to address class imbalance in the decoded multi-label decision space. We compared the performance of the proposed CC k EL algorithm with that of multiple popular multi-label imbalance learning algorithms on 10 benchmark multi-label datasets, and the results show its effectiveness and superiority. [ABSTRACT FROM AUTHOR]

The Bayesian approach to solving inverse problems relies on the choice of a prior. This critical ingredient allows expert knowledge or physical constraints to be formulated in a probabilistic fashion and plays an important role for the success of the inference. Recently, Bayesian inverse problems were solved using generative models as highly informative priors. Generative models are a popular tool in machine learning to generate data whose properties closely resemble those of a given database. Typically, the generated distribution of data is embedded in a low-dimensional manifold. For the inverse problem, a generative model is trained on a database that reflects the properties of the sought solution, such as typical structures of the tissue in the human brain in magnetic resonance imaging. The inference is carried out in the low-dimensional manifold determined by the generative model that strongly reduces the dimensionality of the inverse problem. However, this procedure produces a posterior that does not admit a Lebesgue density in the actual variables and the accuracy attained can strongly depend on the quality of the generative model. For linear Gaussian models, we explore an alternative Bayesian inference based on probabilistic generative models; this inference is carried out in the original high-dimensional space. A Laplace approximation is employed to analytically derive the prior probability density function required, which is induced by the generative model. Properties of the resulting inference are investigated. Specifically, we show that derived Bayes estimates are consistent, in contrast to the approach in which the low-dimensional manifold of the generative model is employed. The MNIST data set is used to design numerical experiments that confirm our theoretical findings. It is shown that the approach proposed can be advantageous when the information contained in the data is high and a simple heuristic is considered for the detection of this case. Finally, the pros and cons of both approaches are discussed. [ABSTRACT FROM AUTHOR]

Continual learning (CL) is a paradigm which addresses the issue of how to learn from sequentially arriving tasks. The goal of this paper is to introduce a CL framework which can both learn from a global multi-task architecture and locally adapt this learning to the task at hand. In addition to the global knowledge, we conjecture that it is also beneficial to further focus on the most relevant pieces of previous knowledge. Using a prototypical network as a proxy, the proposed framework bases its adaptation on the similarity between the current data stream and the previously encountered data. We develop two algorithms, one for the standard task-aware CL and another for the more challenging task-free setting where boundaries between tasks are unknown. We correspondingly derive a generalization upper bound on the error of an upcoming task. Experiments demonstrate that the introduced algorithms lead to improved performance on several CL benchmarks. [ABSTRACT FROM AUTHOR]

Within the federated learning (FL) framework, the client collaboratively trains the model in coordination with a central server, while the training data can be kept locally on the client. Thus, the FL framework mitigates the privacy disclosure and costs related to conventional centralized machine learning. Nevertheless, current surveys indicate that FL still has problems in terms of communication efficiency and privacy risks. In this paper, to solve these problems, we develop an FL framework with communication-efficient and privacy-preserving (FLCP). To realize the FLCP, we design a novel compression algorithm with efficient communication, namely, adaptive weight compression FedAvg (AWC-FedAvg). On the basis of the non-independent and identically distributed (non-IID) and unbalanced data distribution in FL, a specific compression rate is provided for each client, and homomorphic encryption (HE) and differential privacy (DP) are integrated to provide demonstrable privacy protection and maintain the desirability of the model. Therefore, our proposed FLCP smoothly balances communication efficiency and privacy risks, and we prove its security against "honest-but-curious" servers and extreme collusion under the defined threat model. We evaluate the scheme by comparing it with state-of-the-art results on the MNIST and CIFAR-10 datasets. The results show that the FLCP performs better in terms of training efficiency and model accuracy than the baseline method. [ABSTRACT FROM AUTHOR]

Federated learning enables multiple clients to collaboratively learn machine learning models in a privacy-preserving manner. However, in real-world scenarios, a key challenge encountered in federated learning is the statistical heterogeneity among clients. Existing work mainly focused on a single global model shared across the clients, making it hard to generalize well to all clients due to the large discrepancy in the data distributions. To address this challenge, we propose pFedLT, a novel approach that can adapt the single global model to different data distributions. Specifically, we propose to perform a pluggable layer-wise transformation during the local update phase based on scaling and shifting operations. In particular, these operations are learned with a meta-learning strategy. By doing so, pFedLT can capture the diversity of data distribution among clients, therefore, can generalize well even when the data distributions among clients exhibit high statistical heterogeneity. We conduct extensive experiments on synthetic and real-world datasets (MNIST, Fashion_MNIST, CIFAR-10, and Office+Caltech10) under different Non-IID settings. Experimental results demonstrate that pFedLT significantly improves the model accuracy by up to 11.67% and reduces the communication costs compared with state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Introduction: The incidence of breast cancer has steadily risen over the years owing to changes in lifestyle and environment. Presently, breast cancer is one of the primary causes of cancer-related deaths among women, making it a crucial global public health concern. Thus, the creation of an automated diagnostic system for breast cancer bears great importance in the medical community. Objectives: This study analyses the Wisconsin breast cancer dataset and develops a machine learning algorithm for accurately classifying breast cancer as benign or malignant. Methods: Our research is a retrospective study, and the main purpose is to develop a high-precision classification algorithm for benign and malignant breast cancer. To achieve this, we first preprocessed the dataset using standard techniques such as feature scaling and handling missing values. We assessed the normality of the data distribution initially, after which we opted for Spearman correlation analysis to examine the relationship between the feature subset data and the labeled data, considering the normality test results. We subsequently employed the Wilcoxon rank sum test to investigate the dissimilarities in distribution among various breast cancer feature data. We constructed the feature subset based on statistical results and trained 7 machine learning algorithms, specifically the decision tree, stochastic gradient descent algorithm, random forest algorithm, support vector machine algorithm, logistics algorithm, and AdaBoost algorithm. Results: The results of the evaluation indicated that the AdaBoost-Logistic algorithm achieved an accuracy of 99.12%, outperforming the other 6 algorithms and previous techniques. Conclusion: The constructed AdaBoost-Logistic algorithm exhibits significant precision with the Wisconsin breast cancer dataset, achieving commendable classification performance for both benign and malignant breast cancer cases. [ABSTRACT FROM AUTHOR]

Time series forecasting has a wide range of applications in various fields. To eliminate the need for time series data volume, a meta-learning-based few-shot time series forecasting method is proposed. This method uses a residual stack module as its backbone and connects the residuals forward and backward through a multilayer fully connected network so that the model and the meta-learning framework can be seamlessly combined. The Empirical knowledge of different time-sequence tasks is obtained through meta-training. To enable fast adaptation to new prediction tasks, a small meta-network is introduced to adaptively and dynamically generate the learning rate and weight decay coefficient of each step in the network. This method can use sequences of different data distribution characteristics for cross-task learning, and each training task only needs a small number of time series to achieve sequence prediction for the target task. The results show that compared with the two baselines, the proposed method has improved performance on 67.07% and 58.53% of the evaluated tasks. Thus, this method can effectively alleviate the problems caused by insufficient data during training and has broad application prospects in the field of time series. [ABSTRACT FROM AUTHOR]

In order to cope with ever-evolving and increasing cyber threats, intrusion detection systems have become a crucial component of cyber security. Compared with signature-based intrusion detection methods, anomaly-based methods typically employ machine learning techniques to train detection models and possess the capability to discover unknown attacks. However, intrusion detection methods face the challenge of low detection rates for minority class attacks due to imbalanced data distributions. Traditional intrusion detection algorithms address this issue by resampling or generating synthetic data. Additionally, reinforcement learning, as a machine learning method that interacts with the environment to obtain feedback and improve performance, is gradually being considered for application in the field of intrusion detection. This paper proposes a reinforcement-learning-based intrusion detection method that innovatively uses adaptive sample distribution dual-experience replay to enhance a reinforcement learning algorithm, aiming to effectively address the issue of imbalanced sample distribution. We have also developed a reinforcement learning environment specifically designed for intrusion detection tasks. Experimental results demonstrate that the proposed model achieves favorable performance on the NSL-KDD, AWID, and CICIoT2023 datasets, effectively dealing with imbalanced data and showing better classification performance in detecting minority attacks. [ABSTRACT FROM AUTHOR]

Co-clustering is the simultaneous clustering of the samples and attributes of a data matrix that provides deeper insight into data than traditional clustering. However, there is a lack of representation learning algorithms that serve this mechanism of co-clustering, and the current representation learning algorithms are limited to the sample perspective and lack the use of information in the attribute perspective. To solve this problem, in this article, ctSNE, a co-representation learning model based on t-distributed stochastic neighbor embedding, is proposed for unsupervised co-clustering, where ctSNE makes the dataset representation outputted more discriminative of row and column clusters (i.e. co-discrimination). On the basis of t-distributed stochastic neighbor embedding retaining the sample data distribution and local data structure, the philosophy of collaboration is introduced (i.e., row and column hidden relationship information) so that the ctSNE model is equipped with co-representation learning capability, which can effectively improve the performance of co-clustering. To prove the effectiveness of the ctSNE model, several classic co-clustering algorithms are used to check the co-representation performance of ctSNE, and a novel internal index based on an internal clustering index, known as total inertia, is proposed to demonstrate the effect of co-clustering. The numerous experimental results show that ctSNE has tremendous co-representation capability and can significantly improve the performance of co-clustering algorithms. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Harbin University of Science & Technology is the property of Journal of Harbin University of Science & Technology 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.)