Search

Aiming at the problem of inaccurate transformer fault diagnosis in dissolved gas analysis, this paper proposes a novel diagnostic method that integrates an enhanced honey badger algorithm (EHBA) with an ensemble learning-based deep hybrid kernel extreme learning machine (DHKELM). First, kernel principal component analysis (KPCA) was deployed for feature fusion of the gas data, thus extracting more effective features. The DHKELM, combining polynomial and RBF kernel functions, was used as a base learning to build a powerful classifier with Adaboost framework. The EHBA introduces information sharing and firefly perturbation strategies based on HBA. This EHBA was harnessed to optimize the DHKELM's critical parameters, establishing the EHBA-DHKELM-Adaboost transformer fault diagnosis model. Finally, the features garnered by KPCA were fed into the model, simulating and validating various fault diagnosis models. The findings reveal that EHBA-DHKELM-Adaboost achieves 98.75% diagnostic accuracy in transformer faults, surpassing other models. [ABSTRACT FROM AUTHOR]

Gaussian Radial Basis Function Kernels are the most-often-employed kernel function in artificial intelligence for providing the optimal results in contrast to their respective counterparts. However, our understanding surrounding the utilization of the Generalized Gaussian Radial Basis Function across different machine learning algorithms, such as kernel regression, support vector machines, and pattern recognition via neural networks is incomplete. The results delivered by the Generalized Gaussian Radial Basis Function Kernel in the previously mentioned applications remarkably outperforms those of the Gaussian Radial Basis Function Kernel, the Sigmoid function, and the ReLU function in terms of accuracy and misclassification. This article provides a concrete illustration of the utilization of the Generalized Gaussian Radial Basis Function Kernel as mentioned earlier. We also provide an explicit description of the reproducing kernel Hilbert space by embedding the Generalized Gaussian Radial Basis Function as an L 2 − measure, which is utilized in implementing the analysis support vector machine. Finally, we provide the conclusion that we draw from the empirical experiments considered in the manuscript along with the possible future directions in terms of spectral decomposition of the Generalized Gaussian Radial Basis Function. [ABSTRACT FROM AUTHOR]

Support vector machine is one of the most used machine learning algorithms with a comprehensive mathematical infrastructure. The power behind the algorithm is the kernel trick that enables the model to overcome non-linear data distributions by using functions that satisfy the Mercer condition. Undoubtedly, the radial basis function (RBF) is among the most widely used of these functions. The RBF kernel, which has a Gaussian curve, performs local boundary surfaces and supports the generalization capability of the model. In this study, a novel kernel function named Logarithmic Kernel Function (LKF), which has a Gaussian-like curve is presented. The crucial contribution of the LKF is that it can model the dataset better than other similarly shaped kernels in the case of a few training samples (10% and 30%). In the study, 6 classifications and 5 regression sets are handled to compare kernels. Instead of giving the hyperparameters needed by the models manually, they are estimated through the Tree Parzen Estimator, which is based on Sequential Modeling Optimization. This estimation process is repeated 10 times and the average results are obtained. The proposed LKF surpasses the most competitive kernels in various classification and regression tasks. • The novel Gaussian-like Logarithmic Kernel Function (LKF) was proposed. • LKF has been shown to work better than other kernels of similar shape at a small amount of training samples. • For a fair comparison, the parameters of kernels were not determined manually. • Tree Parzen Estimator optimization methods was preferred for parameter selection. • In comparisons made with 6 classification and 6 regression tasks. [ABSTRACT FROM AUTHOR]

A method based on kernel extreme learning machine is proposed for accurate localization and recognition of human bones. In this method, the human skeleton is divided into 16 key nodes based on physiology, and the human skeleton model is formed by connecting lines between the nodes. The sequence images of human actions are taken as input and fed into the input and hidden layers of the extreme learning machine. During the learning and training process, a kernel function is used to construct a kernel matrix to complete weight allocation. The experimental study was conducted on the action sequences of figure skaters, and the results showed that our method achieved an accuracy rate of over 90% in recognizing human bones, with individual sequences achieving an accuracy rate of over 95%, which is superior to convolutional neural network (CNN) and long short-term memory (LSTM) methods. [ABSTRACT FROM AUTHOR]

With the improvement of high-throughput technologies in recent years, large multi-dimensional plant omics data have been produced, and big-data-driven yield prediction research has received increasing attention. Machine learning offers promising computational and analytical solutions to interpret the biological meaning of large amounts of data in crops. In this study, we utilized multi-omics datasets from 156 maize recombinant inbred lines, containing 2496 single nucleotide polymorphisms (SNPs), 46 image traits (i-traits) from 16 developmental stages obtained through an automatic phenotyping platform, and 133 primary metabolites. Based on benchmark tests with different types of prediction models, some machine learning methods, such as Partial Least Squares (PLS), Random Forest (RF), and Gaussian process with Radial basis function kernel (GaussprRadial), achieved better prediction for maize yield, albeit slight difference for method preferences among i-traits, genomic, and metabolic data. We found that better yield prediction may be caused by various capabilities in ranking and filtering data features, which is found to be linked with biological meaning such as photosynthesis-related or kernel development-related regulations. Finally, by integrating multiple omics data with the RF machine learning approach, we can further improve the prediction accuracy of grain yield from 0.32 to 0.43. Our research provides new ideas for the application of plant omics data and artificial intelligence approaches to facilitate crop genetic improvements. [ABSTRACT FROM AUTHOR]

Quantum machine learning endeavors to exploit quantum mechanical effects like superposition, entanglement, and interference to enhance the capabilities of classical machine learning methods. One of the most researched quantum machine learning methodologies presently is the quantum support vector machine (QSVM). Researchers have now developed quantum support vector classifiers and substantiated their potential for accelerating computation and enhancing classification accuracy in practical contexts through experimentation. Nevertheless, the utility of QSVM in regression tasks remains a relatively uncharted territory. In light of this, we introduce a quantum kernel estimation-based quantum support vector regression (QKE-QSVR) model for completing regression tasks. Within this proposed model, classical inputs are encoded as quantum feature vectors using the designed quantum feature map circuit with a variable parameter. The inner product between quantum feature vectors give rise to a trainable quantum kernel, which is subject to optimization through our proposed quantum kernel alignment-based regression (QKAR) algorithm, thereby bolstering the model's predictive accuracy when applied to a specific dataset. Subsequently, the trained quantum kernel is incorporated into the classical support vector regression process to construct the decision function and provide predictions for new data points. In this study, we validate the efficacy of the proposed model using three illustrative examples. Experimental findings underscore that, in comparison to the classical support vector machine model, our proposed model demonstrates superior predictive accuracy. [ABSTRACT FROM AUTHOR]

In this paper, a new machine learning (ML) technique is proposed that uses the fine-tuned version of support vector regression for stock forecasting of time series data. Grid search technique is applied over training dataset to select the best kernel function and to optimize its parameters. The optimized parameters are validated through validation dataset. Thus, the tuning of this parameters to their optimized value not only increases model's overall accuracy but also requires less time and memory. Further, this also minimizes the model from being data overfitted. The proposed method is used to analysis different performance parameters of stock market like up-to-daily and up-to-monthly return, cumulative monthly return, its volatility nature and the risk associated with it. Eight different large-sized datasets are chosen from different domain, and stock is predicted for each case by using the proposed method. A comparison is carried out among the proposed method and some similar methods of same interest in terms of computed root mean square error and the mean absolute percentage error. The comparison reveals the proposed method to be more accurate in predicting the stocks for the chosen datasets. Further, the proposed method requires much less time than its counterpart methods. [ABSTRACT FROM AUTHOR]

This paper explores the potential benefits of quantum coherence and quantum discord in the non-universal quantum computing model called deterministic quantum computing with one qubit (DQC1) in supervised machine learning. We show that the DQC1 model can be leveraged to develop an efficient method for estimating complex kernel functions. We demonstrate a simple relationship between coherence consumption and the kernel function, a crucial element in machine learning. The paper presents an implementation of a binary classification problem on IBM hardware using the DQC1 model and analyzes the impact of quantum coherence and hardware noise. The advantage of our proposal lies in its utilization of quantum discord, which is more resilient to noise than entanglement. [ABSTRACT FROM AUTHOR]

Hypertension is associated with significant cardiovascular morbidity. Blood pressure (BP) control on maintenance hemodialysis (HD) is strongly impacted by volume status. The objective of this study was to assess whether machine learning (ML) is effective in predicting post-HD BP in children and young adults on HD. We collected data on BP, IDWG, pulse, and weights for patients on maintenance HD (> 3 months). Input features included DW, pre-post weight difference, IDWG and pre-HD BP. Seven models were trained and tuned using open-source libraries. Model performance was evaluated using time-series cross-validation on a rolling basis (3–12 month training, 1-day testing). Various regression scores were compared between models. Data for 35 patients (14,375 HD sessions) were analyzed. Extreme gradient boosting (XGB) and vector autoregression with exogenous regressors (VARX) achieved better accuracy in predicting post-dialysis systolic BP than K-nearest neighbor, support vector regression (SVR) with radial basis function kernel and random forest (p < 0.001 for each). The differences in accuracy between XGB, VARX, SVR with linear kernel, random forest and linear regression were not significant. Using clinical parameters, ML models may be useful in predicting post-HD BP, which may help guide DW adjustment and optimizing BP control for maintenance HD patients. [ABSTRACT FROM AUTHOR]

A block valve installed along a trunk natural gas pipeline can automatically shut down when it detects a pressure-drop rate that is over a threshold value for a specified duration, indicating that leakage accidents have occurred. However, current methods have difficulties on identifying the differences among the pressure-drop rate signals caused by small-hole leakage, compressor startup, and block valve emergency shut-down conditions. This usually leads to improperly shutting down the block valve. Based on an optimized support vector machine, a machine-learning method is proposed to recognize the pressure-drop rate signals through the eigenvalue of the pressure-drop curve instead of only using the pressure-drop signal and its duration. Firstly, the singular decomposition method and the support vector machine method are applied to initially classify the pressure-drop rate signals of three operating conditions: the gas leakage, compressor startup, and block valve emergency shut-down conditions. Secondly, a teaching-learning-based optimization method is applied to optimize the penalty factor and kernel function parameter in the support vector machine model. Particularly, the tent chaotic maps and adaptive inertia weight methods are applied to improve the teaching-learning-based optimization method to balance its local and global search performance. Thirdly, an improved teaching-learning-based support vector machine method is established. Finally, 960 sets of simulated pressure-drop signals taken from three trunk natural gas pipelines and pressure-drop rate signals collected from an actual pipeline are applied to verify the accuracy of the proposed model. The results show that the improved teaching-learning-based support vector machine model achieved high classification and recognition accuracy. Specifically, it achieved accuracy rates of 99.4%, 100%, 98.3%, and 100% for the pressure-drop rate signals generated by pipelines A, B, C, and Cangzhou branch in the presence of pipeline leakage (the leak hole diameter is from 25 to 125 mm). Additionally, it demonstrated an average recognition accuracy of 97.42% for the pressure-drop rate signals generated by pipelines under other operating conditions. Through cooperating with Supervisory Control And Data Acquisition system, this method provides a more relevant approach to determine the shutdown conditions of a block valve while preventing mistaken actions. [ABSTRACT FROM AUTHOR]

Cell type identification is one of the fundamental tasks in single-cell RNA sequencing (scRNA-seq) studies. It is a key step to facilitate downstream interpretations such as differential expression, trajectory inference, etc. scRNA-seq data contains technical variations that could affect the interpretation of the cell types. Therefore, gene selection, also known as feature selection in data science, plays an important role in selecting informative genes for scRNA-seq cell type identification. Generally speaking, feature selection methods are categorized into filter-, wrapper-, and embedded-based approaches. From the existing literature, methods from filter- and embedded-based approaches are widely applied in scRNA-seq gene selection tasks. The wrapper-based method that gives promising results in other fields has yet been extensively utilized for selecting gene features from scRNA-seq data; in addition, most of the existing wrapper methods used in this field are clustering instead of classification-based. With a large number of annotated data available today, this study applied a classification-based approach as an alternative to the clustering-based wrapper method. In our work, a quantum-inspired differential evolution (QDE) wrapped with a classification method was introduced to select a subset of genes from twelve well-known scRNA-seq transcriptomic datasets to identify cell types. In particular, the QDE was combined with different machine-learning (ML) classifiers namely logistic regression, decision tree, support vector machine (SVM) with linear and radial basis function kernels, as well as extreme learning machine. The linear SVM wrapped with QDE, namely QDE-SVM, was chosen by referring to the feature selection results from the experiment. QDE-SVM showed a superior cell type classification performance among QDE wrapping with other ML classifiers as well as the recent wrapper methods (i.e., FSCAM, SSD-LAHC, MA-HS, and BSF). QDE-SVM achieved an average accuracy of 0.9559, while the other wrapper methods achieved average accuracies in the range of 0.8292 to 0.8872. [ABSTRACT FROM AUTHOR]

Multi-source online transfer learning uses the tagged data from multiple source domains to enhance the classification performance of the target domain. For unbalanced data sets, a multi-source online transfer learning algorithm that can oversample in the feature spaces of the source domain and the target domain is proposed. The algorithm consists of two parts: oversampling multiple source domains and oversampling online target domains. In the oversampling phase of the source domain, oversampling is performed in the feature space of the support vector machine (SVM) to generate minority samples. New samples are obtained by amplifying the original Gram matrix through neighborhood information in the source domain feature space. In the oversampling phase of the online target domain, minority samples from the current batch search for k-nearest neighbors in the feature space from multiple batches that have already arrived, and use the generated new samples and the original samples in the current batch to train the target domain function together. The samples from the source domain and the target domain are mapped to the same feature space through the kernel function for oversampling, and the corresponding decision function is trained using the data from the source domain and the target domain with relatively balanced class distribution, so as to improve the overall performance of the algorithm. Comprehensive experiments were conducted on four real datasets, and compared to other baseline algorithms on the Office Home dataset, the accuracy improved by 0.0311 and the G-mean value improved by 0.0702. [ABSTRACT FROM AUTHOR]

The progressive damage of kidney function and structure exhibits a high risk of cardiovascular disease and is the leading cause of life-threatening chronic kidney disease (CKD). Early diagnosing is the only way to prevent CKD from worsening. Different data mining and machine learning approaches are implemented for predicting chronic kidney disease; however, the extraction of hidden details from the clinical data is a difficult challenge. To overcome this situation, a new model for accurate prediction of chronic kidney disease is proposed in this article. This work evaluated the proposed predictive model using a CKD dataset. The raw chronic kidney disease dataset is initially preprocessed using data cleaning, data reduction as well as data transformation steps. These steps make the classification system easier to predict diseased and normal classes by enhancing the data quality. The features are then selected using a hybrid flash butterfly optimization algorithm. Here, the weight functions of the kernel soft plus extreme learning machine (KSELM) method are optimally tuned using the hybrid flash butterfly optimization (HFBO) algorithm to increase prediction accuracy. The efficiency of the proposed HFBO-KSELM algorithm is examined using different evaluation measures namely precision, accuracy, and specificity, recall, F1-score, and computation time. The results of the experimental analysis display superior performance of the proposed HFBO-KSELM algorithm over other existing methods, particularly with an accuracy of 97.8%, precision of 97.4%, recall of 96.9%, specificity of 97.5%, F1-score of 97.1%, computational time of 2.3 s, training accuracy of 0.978, testing accuracy of 0.94 and ROC/AUC of 0.97. Finally, the KSELM algorithm accurately predicts and classifies them into benign and malignant classes. The experimental result showed that the proposed HFBO-KSELM algorithm provides the effectiveness and robustness to detect chronic kidney disease. [ABSTRACT FROM AUTHOR]

Defining meaningful feature (molecule) combinations can enhance the study of disease diagnosis and prognosis. However, feature combinations are complex and various in biosystems, and the existing methods examine the feature cooperation in a single, fixed pattern for all feature pairs, such as linear combination. To identify the appropriate combination between two features and evaluate feature combination more comprehensively, this paper adopts kernel functions to study feature relationships and proposes a new omics data analysis method KF- k -TSP. Besides linear combination, KF- k -TSP also explores the nonlinear combination of features, and allows hybridizing multiple kernel functions to evaluate feature interaction from multiple views. KF- k -TSP selects k > 0 top-scoring pairs to build an ensemble classifier. Experimental results show that KF- k -TSP with multiple kernel functions which evaluates feature combinations from multiple views is better than that with only one kernel function. Meanwhile, KF- k -TSP performs better than TSP family algorithms and the previous methods based on conversion strategy in most cases. It performs similarly to the popular machine learning methods in omics data analysis, but involves fewer feature pairs. In the procedure of physiological and pathological changes, molecular interactions can be both linear and nonlinear. Hence, KF- k -TSP, which can measure molecular combination from multiple perspectives, can help to mine information closely related to physiological and pathological changes and study disease mechanism. [ABSTRACT FROM AUTHOR]

A prompt and precise estimation of traffic conditions on the scale of a few minutes by analyzing past data is crucial for establishing an effective intelligent traffic management system. Nevertheless, because of the irregularity and nonlinear features of traffic flow data, developing a prediction model with excellent robustness poses a significant obstacle. Therefore, we propose genetic-search-algorithm-improved kernel extreme learning machine, termed GA-KELM, to unleash the potential of improved prediction accuracy and generalization performance. By substituting the inner product with a kernel function, the accuracy of short-term traffic flow forecasting using extreme learning machines is enhanced. The genetic algorithm evades manual traversal of all possible parameters in searching for the optimal solution. The prediction performance of GA-KELM is evaluated on eleven benchmark datasets and compared with several state-of-the-art models. There are four benchmark datasets from the A1, A2, A4, and A8 highways near the ring road of Amsterdam, and the others are D1, D2, D3, D4, D5, D6, and P, close to Heathrow airport on the M25 expressway. On A1, A2, A4, and A8, the RMSEs of the GA-KELM model are 284.67 vehs/h, 193.83 vehs/h, 220.89 vehs/h, and 163.02 vehs/h, respectively, while the MAPEs of the GA-KELM model are 11.67%, 9.83%, 11.31%, and 12.59%, respectively. The results illustrate that the GA-KELM model is obviously superior to state-of-the-art models. [ABSTRACT FROM AUTHOR]

The paper focuses on the importance of prompt and efficient process fault detection in contemporary manufacturing industries, where product quality and safety protocols are critical. The study compares the efficiencies of two techniques for process fault detection: Kernel Principal Component Analysis (KPCA) and the observer method. Both techniques are applied to observe water volume variation within a hydraulic system comprising three tanks. PCA is an unsupervised learning technique used for dimensionality reduction and pattern recognition. It is an extension of Principal Component Analysis (PCA) that utilizes kernel functions to transform data into higher-dimensional spaces, where it becomes easier to separate classes or identify patterns. In this paper, KPCA is applied to detect faults in the hydraulic system by analyzing the variation in water volume. The observer method originates from control theory and is utilized to estimate the internal states of a system based on its output measurements. It is commonly used in control systems to estimate the unmeasurable or hidden states of a system, which is crucial for ensuring proper control and fault detection. In this study, the observer method is applied to the hydraulic system to estimate the water volume variations within the three tanks. The paper presents a comparative study of these two techniques applied to the hydraulic system. The results show that both KPCA and the observer method perform similarly in detecting faults within the system. This similarity in performance highlights the efficacy of these techniques and their potential adaptability in various fault diagnosis scenarios within modern manufacturing processes. [ABSTRACT FROM AUTHOR]

Traditional survival analysis methods have been widely used in medicine and various fields, but they must satisfy some statistical assumptions. Machine learning methods can overcome this weakness. Three support vector-based machine learning-survival analysis models are proposed with an actual cancer dataset. Experiments show that some machine learning models outperform the Cox statistical model. Accurate kernel complexity experiments are designed for different models, and the selected kernel functions are Poly, neural tangent kernel (NTK) and so on. The NTK kernel model with the highest accuracy is investigated in more depth, including computational complexity and robustness. Some interesting properties are found in these aspects that may provide new ideas for researchers. [ABSTRACT FROM AUTHOR]

This paper employs two data-driven methods, Random Forest (RF) and Support Vector Machines (SVM), to develop mineral prospectivity models for an epithermal Au deposit. Four distinct models are presented for comparison: one employing RF and three using SVM with different kernel functions—namely linear, Radial Basis Function (RBF), and polynomial. The analysis leverages a compact training dataset, encompassing just 20 deposits, with deposit and non-deposit locations chosen from known mineral occurrences. Fourteen predictor maps are constructed based on the available data and the exploration model. The findings indicate that RF is more stable and robust than SVM, regardless of the kernel function implemented. While all SVM models outperformed the RF model in terms of classification capability on the training dataset achieving an accuracy exceeding 89% versus 78% for the RF model, the success rate curves suggest superior predictive abilities of RF over SVM models. This implies that the SVM models may be overfitting the training data due to the limited quantity of training deposits. [ABSTRACT FROM AUTHOR]

This paper presents a method for solving linear Fredholm integro-differential equations using a feedforward neural network based on Legendre polynomials. Firstly, the Legendre polynomials are used to approximate the unknown function and the kernel function in the equations. Secondly, The roots of Legendre polynomials are obtained by using Newton iteration method to obtain Gaussian integration points, and the obtained Gaussian integration points are used as the input nodes of the neural network, and the corresponding weight is learned by the gradient descent method to obtain an approximate solution. Finally, through the numerical value Case analysis to verify the effectiveness of the method. [ABSTRACT FROM AUTHOR]

Molecular dynamics (MD) has served as a powerful tool for designing materials with reduced reliance on laboratory testing. However, the use of MD directly to treat the deformation and failure of materials at the mesoscale is still largely beyond reach. In this work, we propose a learning framework to extract a peridynamics model as a mesoscale continuum surrogate from MD simulated material fracture data sets. Firstly, we develop a novel coarse-graining method, to automatically handle the material fracture and its corresponding discontinuities in the MD displacement data sets. Inspired by the weighted essentially non-oscillatory (WENO) scheme, the key idea lies at an adaptive procedure to automatically choose the locally smoothest stencil, then reconstruct the coarse-grained material displacement field as the piecewise smooth solutions containing discontinuities. Then, based on the coarse-grained MD data, a two-phase optimization-based learning approach is proposed to infer the optimal peridynamics model with damage criterion. In the first phase, we identify the optimal nonlocal kernel function from the data sets without material damage to capture the material stiffness properties. Then, in the second phase, the material damage criterion is learnt as a smoothed step function from the data with fractures. As a result, a peridynamics surrogate is obtained. As a continuum model, our peridynamics surrogate model can be employed in further prediction tasks with different grid resolutions from training, and hence allows for substantial reductions in computational cost compared with MD. We illustrate the efficacy of the proposed approach with several numerical tests for the dynamic crack propagation problem in a single-layer graphene. Our tests show that the proposed data-driven model is robust and generalizable, in the sense that it is capable of modeling the initialization and growth of fractures under discretization and loading settings that are different from the ones used during training. [ABSTRACT FROM AUTHOR]

Rapid and nondestructive recognition method of tire rubber is reported by attenuated total reflectance–Fourier transform infrared (ATR–FTIR) spectroscopy with machine learning algorithms. The infrared spectra were obtained from 187 truck, sedan, motorbike, and passenger car tires. Weighted k nearest neighbor analysis (WKNN), support vector machine (SVM), random forest (RF), and logistic regression (LR) were used and compared for modeling. The influences of the k value, kernel function, number of estimators, maximum depth, minimum sample leaf, and other parameters affecting recognition performance were optimized. The results show that the recognition performance of four models was SVM > LR > WKNN > RF. The SVM model (polynomial kernel function, C = 20, gamma = 1) was considered to be optimal. All samples had a recognition accuracy of 100% based on the tire type with a training set value of 96.9% and test set value of 92.3%. Hence, the developed procedure is suitable for the characterization of tires. [ABSTRACT FROM AUTHOR]

One of the most important problems for farmers who produce large amounts of apples is the classification of the apples according to their types in a short time without handling them. Support vector machines (SVM) and deep residual networks (ResNet-50) are machine learning methods that are able to solve general classification situations. In this study, the classification of apple varieties according to their genus is made using machine learning algorithms. A database is created by capturing 120 images from six different apple species. Bag of visual words (BoVW) treat image features as words representing a sparse vector of occurrences over the vocabulary. BoVW features are classified using SVM. On the other hand, ResNet-50 is a convolutional neural network that is 50 layers deep with embedded feature extraction layers. The pre-trained ResNet-50 architecture is retrained for apple classification using transfer learning. In the experiments, our dataset is divided into three cases: Case 1: 40% train, 60% test; Case 2: 60% train, 40% test; and Case 3: 80% train, 20% test. As a result, the linear, Gaussian, and polynomial kernel functions used in the BoVW + SVM algorithm achieved 88%, 92%, and 96% accuracy in Case 3, respectively. In the ResNet-50 classification, the root-mean-square propagation (rmsprop), adaptive moment estimation (adam), and stochastic gradient descent with momentum (sgdm) training algorithms achieved 86%, 89%, and 90% accuracy, respectively, in the set of Case 3. [ABSTRACT FROM AUTHOR]

The discovery of quantum algorithms offering provable advantages over the best known classical alternatives, together with the parallel ongoing revolution brought about by classical artificial intelligence, motivates a search for applications of quantum information processing methods to machine learning. Among several proposals in this domain, quantum kernel methods have emerged as particularly promising candidates. However, while some rigorous speedups on certain highly specific problems have been formally proven, only empirical proof-of-principle results have been reported so far for real-world datasets. Moreover, no systematic procedure is known, in general, to fine tune and optimize the performances of kernel-based quantum classification algorithms. At the same time, certain limitations such as kernel concentration effects—hindering the trainability of quantum classifiers—have also been recently pointed out. In this work, we propose several general-purpose optimization methods and best practices designed to enhance the practical usefulness of fidelity-based quantum classification algorithms. Specifically, we first describe a data pre-processing strategy that, by preserving the relevant relationships between data points when processed through quantum feature maps, substantially alleviates the effect of kernel concentration on structured datasets. We also introduce a classical post-processing method that, based on standard fidelity measures estimated on a quantum processor, yields non-linear decision boundaries in the feature Hilbert space, thus achieving the quantum counterpart of the radial basis functions technique that is widely employed in classical kernel methods. Finally, we apply the so-called quantum metric learning protocol to engineer and adjust trainable quantum embeddings, demonstrating substantial performance improvements on several paradigmatic real-world classification tasks. [ABSTRACT FROM AUTHOR]

Carbon trading has been deemed as the most effective mechanism to mitigate carbon emissions. However, during carbon trading market operation, competition among market participants will inevitably occur; hence, the precise forecasting of the carbon trading price (CTP) has become a significant element in the formulation of competition strategies. This investigation has established a hybrid CTP forecasting framework combining complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), sample entropy (SE) method, improved salp swarm algorithm (ISSA), and multi-kernel extreme learning machine (MKELM) methods to improve forecasting accuracy. Firstly, the initial CTP data sequence is disintegrated into several intrinsic mode functions (IMFs) and a residual sequence by a CEEMDAN method. Secondly, to save calculation time, SE method has been utilized to reconstruct the IMFs and the residual sequence into new IMFs. Thirdly, the new IMFs are fed into the MKELM model, combing RBF and the poly kernel functions to utilize their superior learning and generalization abilities. The parameters of the MKELM model are optimized by ISSA, combining dynamic inertia weight and chaotic local searching method into the SSA to enhance the searching speed, convergence precision, as well as the global searching ability. CTP data in Guangdong, Shanghai, and Hubei are selected to prove the validity of the established CEEMDAN-SE-ISSA-MKELM model. Through a comparison analysis, the established CEEMDAN-SE-ISSA-MKELM model performs the best with the smallest MAPE and RMSE values and the highest R2 value, which are 0.76%, 0.53, and 0.99, respectively, for Guangdong,. Thus, the presented model would be extensively applied in CTP forecasting in the future. [ABSTRACT FROM AUTHOR]

Predicting the functional properties of many molecular systems relies on understanding how atomistic interactions give rise to macroscale observables. However, current attempts to develop predictive models for the structural and thermodynamic properties of condensed-phase systems often rely on extensive parameter fitting to empirically selected functional forms whose effectiveness is limited to a narrow range of physical conditions. In this article, we illustrate how these traditional fitting paradigms can be superseded using machine learning. Specifically, we use the results of molecular dynamics simulations to train machine learning protocols that are able to produce the radial distribution function, pressure, and internal energy of a Lennard-Jones fluid with increased accuracy in comparison to previous theoretical methods. The radial distribution function is determined using a variant of the segmented linear regression with the multivariate function decomposition approach developed by Craven et al. [J. Phys. Chem. Lett. 11, 4372 (2020)]. The pressure and internal energy are determined using expressions containing the learned radial distribution function and also a kernel ridge regression process that is trained directly on thermodynamic properties measured in simulation. The presented results suggest that the structural and thermodynamic properties of fluids may be determined more accurately through machine learning than through human-guided functional forms. [ABSTRACT FROM AUTHOR]

By mimicking the hierarchical structure of human brain, deep spiking neural networks (DSNNs) can extract features from a lower level to a higher level gradually, and improve the performance for the processing of spatio-temporal information. Due to the complex hierarchical structure and implicit nonlinear mechanism, the formulation of spike train level supervised learning methods for DSNNs remains an important problem in this research area. Based on the definition of kernel function and spike trains inner product (STIP) as well as the idea of error backpropagation (BP), this paper firstly proposes a deep supervised learning algorithm for DSNNs named BP-STIP. Furthermore, in order to alleviate the intrinsic weight transport problem of the BP mechanism, feedback alignment (FA) and broadcast alignment (BA) mechanisms are utilized to optimize the error feedback mode of BP-STIP, and two deep supervised learning algorithms named FA-STIP and BA-STIP are also proposed. In the experiments, the effectiveness of the proposed three DSNN algorithms is verified on the MNIST digital image benchmark dataset, and the influence of different kernel functions on the learning performance of DSNNs with different network scales is analyzed. Experimental results show that the FA-STIP and BP-STIP algorithms can achieve 94.73% and 95.65% classification accuracy, which apparently possess better learning performance and stability compared with the benchmark algorithm BP-STIP. [ABSTRACT FROM AUTHOR]

The information detection of complex systems from data is currently undergoing a revolution, driven by the emergence of big data and machine learning methodology. Discovering governing equations and quantifying the dynamical properties of complex systems are among the central challenges. In this work, we devised a nonparametric approach to learning the relative entropy rate from observations of stochastic differential equations with different drift functions. The estimator corresponding to the relative entropy rate is then presented via the Gaussian process kernel theory. Meanwhile, this approach enables us to extract the governing equations. We illustrate our approach with several examples. Numerical experiments show the proposed approach performs well for rational drift functions, not only polynomial drift functions. [ABSTRACT FROM AUTHOR]

• A novel classification method named projection twin extreme learning machine (PTELM) is proposed. • A new method termed as capped L1-norm projection twin extreme learning machine (CL1-PTELM) is proposed by introducing capped L1-norm distance metric into PTELM. • Two efficient algorithms are designed for our PTELM and CL1-PTELM, and two theorems are proved for CL1-PTELM's convergence and local optimality. • Sufficient experiments on multiple datasets demonstrate the effectiveness of the proposed methods. In this paper, we incorporate the idea of projection twin support vector machines (PTSVM) into the basic framework of twin extreme learning machines (TELM) and first propose a novel binary classifier named projection twin extreme learning machines (PTELM). PTELM is to seek two projection directions in the TELM feature space, such that the projected samples of one class are well separated from those of the other class. Compared with the PTSVM, PTELM tackles nonlinear cases without using several fixed kernel functions, thus PTELM is less sensitive to use specified parameters and can get better classification accuracy. Then, a new capped L 1 -norm PTELM (C L 1 -PTELM) is proposed by introducing capped L 1 -norm distance metric in PTELM to reduce the effect of outliers. C L 1 -PTELM overcomes the disadvantages of L 2 -norm distance metric and hinge loss. Thus, C L 1 -PTELM enhances the robust performance of our PTELM. Finally, two effective algorithms are designed to solve the problem of PTELM and to deal with the challenge of C L 1 -PTELM brought by non-convex optimization problem, respectively. Simultaneously, we theoretically prove the convergence and local optimality of C L 1 -PTELM algorithm. Numerical experiments on three synthetic datasets and several UCI datasets show the feasibility and effectiveness of our proposed methods. [ABSTRACT FROM AUTHOR]

Tail biting is a damaging behaviour that impacts the welfare and health of pigs. Early detection of precursor signs of tail biting provides the opportunity to take preventive measures, thus avoiding the occurrence of the tail biting event. This study aimed to build a machine-learning algorithm for real-time detection of upcoming tail biting outbreaks, using feeding behaviour data recorded by an electronic feeder. Prediction capacities of seven machine learning algorithms (Generalized Linear Model with Stepwise Feature Selection, random forest, Support Vector Machines with Radial Basis Function Kernel, Bayesian Generalized Linear Model, Neural network, K-nearest neighbour, and Partial Least Squares Discriminant Analysis) were evaluated from daily feeding data collected from 65 pens originating from two herds of grower-finisher pigs (25-100kg), in which 27 tail biting events occurred. Data were divided into training and testing data in two different ways, either by randomly splitting data into 75% (training set) and 25% (testing set), or by randomly selecting pens to constitute the testing set. In the first data splitting, the model is regularly updated with previous data from the pen, whereas in the second data splitting, the model tries to predict for a pen that it has never seen before. The K-nearest neighbour algorithm was able to predict 78% of the upcoming events with an accuracy of 96%, when predicting events in pens for which it had previous data. Our results indicate that machine learning models can be considered for implementation into automatic feeder systems for real-time prediction of tail biting events. [ABSTRACT FROM AUTHOR]

In this paper, we propose using a propeller modulation on the transmitted signal (called sonar micro-Doppler) and different support vector machine (SVM) kernels for automatic recognition of moving sonar targets. In general, the main challenge for researchers and craftsmen working in the field of sonar target recognition is the lack of access to a valid and comprehensive database. Therefore, using a comprehensive mathematical model to simulate the signal received from the target can respond to this challenge. The mathematical model used in this paper simulates the return signal of moving sonar targets well. The resulting signals have unique properties and are known as frequency signatures. However, to reduce the complexity of the model, the 128-point fast Fourier transform (FFT) is used. The selected SVM classification is the most popular machine learning algorithm with three main kernel functions: RBF kernel, linear kernel, and polynomial kernel tested. The accuracy of correctly recognizing targets for different signal-to-noise ratios (SNR) and different viewing angles was assessed. Accuracy detection of targets for different SNRs (-20, -15, -10, -5, 0, 5, 10, 15, 20) and different viewing angles (10, 20, 30, 40, 50, 60, 70, 80) is evaluated. For a more fair comparison, multilayer perceptron neural network with two back-propagation (MLP-BP) training methods and gray wolf optimization (MLP-GWO) algorithm were used. But unfortunately, considering the number of classes, its performance was not satisfactory. The results showed that the RBF kernel is more capable for high SNRs (SNR = 20, viewing angle = 10) with an accuracy of 98.528%. [ABSTRACT FROM AUTHOR]

Leaf area index and dry matter mass are important indicators for crop growth and yields. In order to solve the problem of predicting the summer maize growth index and yield under different soil quality and field management conditions, this study proposes a prediction model based on the supervised machine learning regression algorithm. Firstly, the data pool was constructed by collecting the measured data for maize in the main planting area. The total water input (rainfall plus irrigation water), fertilization, soil quality, and planting density were selected as the training set. Then, the maximum leaf area index (LAImax), maximum dry material mass (Dmax), and summer maize yields (Y) in the data pool were trained by using Gaussian regression (rational quadratic kernel function and Matern kernel function), support vector machine (SVM) and linear regression models. The training models were verified with the data-set not included in the data pool, and the water and fertilizer coupling functions were developed. The prediction results showed that compared to the support vector machine models and the linear regression models, the Gaussian regression prediction models comprising the rational quadratic and Matern kernel functions had good prediction accuracy. The coefficients of determination (R2) of the prediction results were 0.91, 0.89 and 0.88; the root-mean-square errors (RMSEs) were 0.3, 1138.6 and 666.16 kg/hm2; and the relative root-mean-square errors (rRMSEs) were 6.3%, 5.94% and 6.53% for LAImax, Dmax and Y, respectively. The optimal total water inputs and nitrogen applications indicated by the prediction results and the water and fertilizer coupling functions were consistent with the measured range from the field tests. The supervised machine learning regression algorithm provides a simple method to predict the yield of maize and optimize the total water inputs and nitrogen applications using only the soil quality and planting density. [ABSTRACT FROM AUTHOR]

To explore the application of traditional machine learning model in the intelligent management of pigs, in this paper, the influence of PCA pre-treatment on pig face identification with SVM is studied. By testing method, the kernel functions of two testing schemes, one adopting SVM alone and the other adopting PCA+SVM, were determined to be poly and RBF, whose coefficients were 0.03 and 0.01, respectively. With individual identification tests carried out on 10 pigs respectively, the identification accuracy was increased to 88.85% from 83.66% by the improved scheme, also the training time as well as testing time were reduced to 30.1% and 20.97% of the original value in the earlier scheme, respectively. It indicates that PCA pre-treatment had a positive effect on improving the efficiency of individual pig identification with SVM. It provides experimental support for the mobile terminals and embedded application of SVM classifiers. [ABSTRACT FROM AUTHOR]

This study reported the causes of prediction uncertainty in landslide susceptibility, based on machine learning. The landslide inventory and conditioning factors were first prepared. Thereafter, factor analysis, including the effectiveness, multicollinearity, and importance was performed for the subsequent susceptibility modeling. Finally, the predicted capacity was explored in terms of model performance and map performance. Investigation indicates that landslides are focused on the central and southeastern part of the study area, displaying the characteristics of small-scale and shallow sliding. The difference in the ML algorithm, kernel functions, the selection of base-classifier, the evaluation indicators of performance, and even the factor zonation results in difficulties in determining the optimal ML model. Besides, the DT-Boosting keeps the high AUC value, 0.981 and 0.862 respectively in the training and test phases, while the RF holds a highly robust in terms of fluctuation characteristics. Indeed, the susceptibility map generated by RBF-ANN can better capture the difference in landslide susceptibility, with the D-value of 7.9. This study provides an in-depth understanding of the uncertainty of the predicted capacity in ML-based landslide susceptibility. [ABSTRACT FROM AUTHOR]

Chronic kidney disease is a gradual loss of kidney function. Determining the important attributes that describe this disease plays a key role in screening and examining the disease by field specialists. The main aim of this study is to comprehensively compare the attribute selection algorithms for predicting this disease. With this aim, several models were built and compared using well-known performance metrics such as accuracy, sensitivity, and specificity in the experiments. Two different attribute selection methods; the stability selection and the minimum redundancy maximum relevance were compared comprehensively on the unbalanced and balanced datasets. In this framework, the stability selection method gave the important attributes. The support vector machines with radial bases function kernel successfully performed the classification using these attributes for this problem. [ABSTRACT FROM AUTHOR]

Virtual screening has significantly improved the success rate of early stage drug discovery. Recent virtual screening methods have improved owing to advances in machine learning and chemical information. Among these advances, the creative extraction of drug features is important for predicting drug–target interaction (DTI), which is a large-scale virtual screening of known drugs. Herein, we report Kullbeck–Leibler divergence (KLD) as a DTI feature and the feature-driven classification model applicable to DTI prediction. For the purpose, E3FP three-dimensional (3D) molecular fingerprints of drugs as a molecular representation allow the computation of 3D similarities between ligands within each target (Q–Q matrix) to identify the uniqueness of pharmacological targets and those between a query and a ligand (Q–L vector) in DTIs. The 3D similarity matrices are transformed into probability density functions via kernel density estimation as a nonparametric estimation. Each density model can exploit the characteristics of each pharmacological target and measure the quasi-distance between the ligands. Furthermore, we developed a random forest model from the KLD feature vectors to successfully predict DTIs for representative 17 targets (mean accuracy: 0.882, out-of-bag score estimate: 0.876, ROC AUC: 0.990). The method is applicable for 2D chemical similarity. [ABSTRACT FROM AUTHOR]

FAO Blaney-Criddle has been generally an accepted method for estimating reference crop evapotranspiration. In this regard, it is inevitable to estimate the b-factor provided by the Food and Agriculture Organization (FAO) of the United Nations Irrigation and Drainage Paper number 24. In this study, five soft computing methods, namely random forest (RF), M5 model tree (M5), support vector regression with the polynomial function (SVR-poly), support vector regression with radial basis function kernel (SVR-rbf), and random tree (RT), were adapted to estimate the b-factor. And Their performances were also compared. The suitable hyper-parameters for each soft computing method were investigated. Five statistical indices were deployed to evaluate their performance, i.e., the coefficient of determination (r2), the mean absolute relative error (MARE), the maximum absolute relative error (MXARE), the standard deviation of the absolute relative error (DEV), and the number of samples with an error greater than 2% (NE > 2%). Findings reveal that SVR-rbf gave the highest performance among five soft computing models, followed by the M5, RF, SVR-poly, and RT. The M5 also derived a new explicit equation for b estimation. SVR-rbf provided a bit lower efficacy than the radial basis function network but outperformed the regression equations. Models' Applicability for estimating monthly reference evapotranspiration (ETo) was demonstrated. [ABSTRACT FROM AUTHOR]

Some response surface functions in complex engineering systems are usually highly nonlinear, unformed, and expensive to evaluate. To tackle this challenge, Bayesian optimization (BO), which conducts sequential design via a posterior distribution over the objective function, is a critical method used to find the global optimum of black-box functions. Kernel functions play an important role in shaping the posterior distribution of the estimated function. The widely used kernel function, e.g., radial basis function (RBF), is very vulnerable and susceptible to outliers; the existence of outliers is causing its Gaussian process (GP) surrogate model to be sporadic. In this article, we propose a robust kernel function, asymmetric elastic net radial basis function (AEN-RBF). Its validity as a kernel function and computational complexity are evaluated. When compared with the baseline RBF kernel, we prove theoretically that AEN-RBF can realize smaller mean squared prediction error under mild conditions. The proposed AEN-RBF kernel function can also realize faster convergence to the global optimum. We also show that the AEN-RBF kernel function is less sensitive to outliers, and hence improves the robustness of the corresponding BO with GPs. Through extensive evaluations carried out on synthetic and real-world optimization problems, we show that AEN-RBF outperforms the existing benchmark kernel functions. Note to Practitioners—Some industrial systems cannot be accurately represented by physical models. In this situation, data-driven black-box optimization is necessary for advancing the system automation and intelligence. BO is one of the widely used strategies for learning the global optimum of black-box functions. BO has been applied to robotics, anomaly detection, automatic learning algorithm configuration, reinforcement learning, and deep learning. This article proposes one new kernel function, named after AEN-RBF. The new kernel function will make BO with GPs more robust to outliers and lower the data quality barrier of model training. This article was motivated by the hyperparameter tuning problem of deep learning models for image defect detection in advanced manufacturing, but the method can be easily extended to other applications where kernel functions are needed. Our proposed method is verified by synthetic and real-world optimization problems. [ABSTRACT FROM AUTHOR]

The classification of imbalanced datasets is one of the main challenges in machine learning techniques. Support vector machine (SVM), which generates a model biased within the majority class, usually has bad performance on the minority class because this class may be considered incorrectly as noises. Moreover, datasets often include noises and outliers, and SVM cannot effectively deal with those datasets. In this paper, to defeat the aforementioned challenges, we propose intuitionistic fuzzy twin support vector machines for imbalanced data (IFTSVM-ID). The proposed method can easily handle imbalanced datasets in the presence of noises and outliers. A reasonable weighting strategy is offered to deal with imbalanced classes, and a margin-based technique is assigned to reduce the impact of noise and outliers. We formulate the linear and non-linear kernel functions to find two non-parallel hyperplanes. One real-world and thirty-two imbalanced datasets are selected to validate the performance of IFTSVM-ID. The Friedman test and the bootstrap technique with 95% confidence interval are applied to quantify the results statistically. The experimental results show that our proposed method has much better performance in comparison with other similar techniques. [ABSTRACT FROM AUTHOR]

Data-driven prediction of molecular properties presents unique challenges to the design of machine learning methods concerning data structure/dimensionality, symmetry adaption, and confidence management. In this paper, we present a kernel-based pipeline that can learn and predict the atomization energy of molecules with high accuracy. The framework employs Gaussian process regression to perform predictions based on the similarity between molecules, which is computed using the marginalized graph kernel. To apply the marginalized graph kernel, a spatial adjacency rule is first employed to convert molecules into graphs whose vertices and edges are labeled by elements and interatomic distances, respectively. We then derive formulas for the efficient evaluation of the kernel. Specific functional components for the marginalized graph kernel are proposed, while the effects of the associated hyperparameters on accuracy and predictive confidence are examined. We show that the graph kernel is particularly suitable for predicting extensive properties because its convolutional structure coincides with that of the covariance formula between sums of random variables. Using an active learning procedure, we demonstrate that the proposed method can achieve a mean absolute error of 0.62 ± 0.01 kcal/mol using as few as 2000 training samples on the QM7 dataset. [ABSTRACT FROM AUTHOR]

This paper demonstrates the applicability of machine learning algorithms in sand production problems with natural gas hydrate (NGH)-bearing sands, which have been regarded as a grave concern for commercialization. The sanding problem hinders the commercial exploration of NGH reservoirs. The common sand production prediction methods need assumptions for complicated mathematical derivations. The main contribution of this paper was to introduce machine learning into the prediction sand production by using data from laboratory experiments. Four main machine learning algorithms were selected, namely, K-Nearest Neighbor, Support Vector Regression, Boosting Tree, and Multi-Layer Perceptron. Training datasets for machine learning were collected from a sand production experiment. The experiment considered both the geological parameters and the sand control effect. The machine learning algorithms were mainly evaluated according to their mean absolute error and coefficient of determination. The evaluation results showed that the most accurate results under the given conditions were from the Boosting Tree algorithm, while the K-Nearest Neighbor had the worst prediction performance. Considering an ensemble prediction model, the Support Vector Regression and Multi-Layer Perceptron could also be applied for the prediction of sand production. The tuning process revealed that the Gaussian kernel was the proper kernel function for improving the prediction performance of SVR. In addition, the best parameters for both the Boosting Tree and Multi-Layer Perceptron were recommended for the accurate prediction of sand production. This paper also involved one case study to compare the prediction results of the machine learning models and classic numerical simulation, which showed the capability of machine learning of accurately predicting sand production, especially under stable pressure conditions. [ABSTRACT FROM AUTHOR]

Support vector machine (SVM) is a popular and important machine learning technique based on the central idea of maximizing the minimummargin , i.e., the smallest distance from the training points to the classification boundary. However, recent theoretical studies revealed that maximizing the minimum is not crucial to the generalization performance, while the margindistribution is essential. Furthermore, kernel-free quadratic support vector machine becomes a new trend in the SVM area since it can avoid some difficulties such as finding appropriate kernel functions and tuning their hyper-parameters. This paper proposes two novel non-kernel quadratic surface optimal margin distribution machine (QSODM L and QSODM) models. Their basic idea is to optimize the margin distribution of training points by maximizing the margin mean, minimizing the margin variance and classifying data points by directly constructing a quadratic surface in the original space. These proposed models are convex so that they can be solved by some well-known solvers. Moreover, in order to improve the efficiencies of proposed models on large-scale data sets, a stochastic gradient descent with variance reduction (SVRG) algorithm is proposed to solve the proposed models. The numerical results on some artificial and benchmark data sets verify that QSODM L and QSODM are superior to some other well-known SVM models in classification accuracy and computational efficiency. [ABSTRACT FROM AUTHOR]

Path tracking error control is an essential functionality in the development of autonomous vehicles to follow a planned trajectory. Significant path tracking errors could lead to a collision or even out of the control of the vehicle. Model-based control strategies have been developed to minimize the vehicle's path tracking errors. However, the vehicle model may not truly represent the actual vehicle dynamics. Furthermore, the parameters employed in the vehicle dynamic model may not represent the actual operating conditions of the vehicle under environmental uncertainty. This paper proposes a real-time bias-learning method coupling with the model predictive control (MPC) to improve the fidelity of a baseline vehicle model with the aid of a few experiments (or virtual experiments) so that the path tracking error can be reduced in real-time operation. Gaussian process (GP) regression and recurrent neural network (RNN) are employed for bias-learning and their effectiveness are compared under different scenarios. GP regression learns non-linearity of the model bias through its nonlinear kernel function, whereas the RNN model formulates the bias as a linear combination of hidden nodes which capture the non-linearity of the model bias with a recurrent form. Results reveal that RNN is more effective for real-time learning of the nonlinear model bias than the classical GP regression and the proposed bias-learning model is able to improve the fidelity of a baseline vehicle dynamic model. Consequently, path tracking performance can be greatly improved under environmental uncertainty using the bias-learning-based MPC. [ABSTRACT FROM AUTHOR]

The inappropriate distribution of samples is known as the data imbalance problem leading to minority and majority classes, often in the healthcare domain due to weakly labelled data. The challenge of working with these imbalanced datasets is to develop predictive models by augmenting the samples to minority class to equalize to the number of samples of the majority class. The successful integration of double radial basis function kernel with deep sampling and convolutional neural networks is experimented to propose a hybrid sampling method coined as DRBF-DS, which generates extended samples, and augments those to the minority class in this study. In this paper, the proposed DRBF-DS is compared with the variants of simple and widely used probability and non-probability-based sampling strategies. It confirms that the proposed approach outperforms the sampling strategies in terms of accuracy and other performance measures used for validation. Experimental validations are performed on five publicly available complex imbalanced datasets to demonstrate the effectiveness of the proposed DRBF-DS based on various evaluation metrics. [ABSTRACT FROM AUTHOR]

Dyslexia is among the most common neurological disorders in children. Detection of dyslexia therefore remains an important pursuit for the research works across various domains which is illustrated by the plethora of work presented in diverse scientific articles. The work presented herein attempted to utilize the potential of a unified gaming test of subjects (dyslexia/controls) in tandem with principal components derived from data to detect dyslexia. The work aims to build a machine learning model for dyslexia detection using comprehensive gaming test data. We have attempted to explore the potential of various kernel functions of the Support Vector Machine (SVM) on different number of principal components to reduce the computational complexity. A detection accuracy of 92% is obtained from the radial basis function with 5 components, and the highest detection accuracy obtained from the radial basis function with 3 components is 93%. On the contrary, the Artificial Neural Network(ANN) shows an added advantage with minimal number of hyperparameters with 3 components for obtaining an accuracy of 95%. The comparison of the proposed method with some of the existing works shows efficacy of this method for dyslexia detection. [ABSTRACT FROM AUTHOR]

St37 steel is a frequently preferred material in many structural elements. Parts with low surface roughness values are preferred in terms of fatigue, wear, and corrosion. Since the experimental determination of surface roughness is timeconsuming and costly, these situations can be eliminated by predicting the surface roughness with machine learning methods. In this study, St37 steel was machined in different parameters on a CNC lathe. The surface roughness (Ra, µm) of the experimental samples was measured separately for different process parameters with a profilometer, and 72 different data were obtained. The experimental values obtained were modeled using support vector regression (SVR). The SVR is modeled using two different core functions. These are linear and Gaussian kernel functions. Two different conditions (dry and wet conditions), spindle speed (rpm) and feed rate (mm/rev) were used as input parameters in the models. The output parameter is the surface roughness (Ra, µm). As a result of the designed models, it has been seen that the SVR using the Gaussian kernel function performs better. In this model, the R-square value was obtained as 0.8876, while it was obtained as 0.7231 in the DVR with linear function. Time, material, and cost losses can be reduced by using the SVR with Gaussian core function in industries producing parts from St37 steel by turning. [ABSTRACT FROM AUTHOR]

Support vector machine (SVM) is one of the most important class of machine learning models and algorithms, and has been successfully applied in various fields. Nonlinear optimization plays a crucial role in SVM methodology, both in defining the machine learning models and in designing convergent and efficient algorithms for large-scale training problems. In this paper we present the convex programming problems underlying SVM focusing on supervised binary classification. We analyze the most important and used optimization methods for SVM training problems, and we discuss how the properties of these problems can be incorporated in designing useful algorithms. [ABSTRACT FROM AUTHOR]

• An interpretable machine learning method is developed for the SPAC system. • A local Gaussian process group is built to simulate multiple SPAC subprocesses. • Improved kernel function can interpret the key mechanisms learned by machine. • Improved kernel function can capture nonstationary dynamics during crop growth. • Our method leads to advanced prediction accuracy and generalization ability. Traditional models of the soil-plant-atmosphere continuum (SPAC) system are physics-based and their practical application has been hindered by issues of high parameterization, prior structural bias, and costly running. In this paper, we developed a machine learning modelling method based on the Gaussian process (GP) to avoid these difficulties of traditional modelling. At the same time, shortcomings of conventional machine learning in terms of interpretability and mechanism representation were addressed. The innovative method consisted of structuring a local GP group framework and improving kernel functions used in each local GP. The resultant model was endowed with advanced capacities, including representing interplays between subprocesses within the complex SPAC system, representing nonstationary subprocess dynamics caused by crop growth stage shifts, as well as automatically interpreting key low-order input interactions and dominant input variables for each subprocess. The performance of our model was examined on synthetic SPAC system datasets that covered three different soil conditions. Results demonstrated that its interpretations regarding subprocess mechanisms were robust across different soil conditions and consistent with domain knowledge. Compared to a conventional global GP model and two deep learning models (DNN and LSTM), our model performed significantly better not only in regular prediction experiments but also in generalization experiments that required models to be transferred across different conditions. Our study suggested that it is promising to enable a machine learning model interpretable by improving its feature representation function. For machine learning modelling of complex systems like the SPAC, it may be more important to enhance the model's mechanism representation capability than simply using deeper networks. [ABSTRACT FROM AUTHOR]

We assess Gaussian process (GP) regression as a technique to model interatomic forces in metal nanoclusters by analyzing the performance of 2-body, 3-body, and many-body kernel functions on a set of 19-atom Ni cluster structures. We find that 2-body GP kernels fail to provide faithful force estimates, despite succeeding in bulk Ni systems. However, both 3- and many-body kernels predict forces within an ∼0.1 eV/Å average error even for small training datasets and achieve high accuracy even on out-of-sample, high temperature structures. While training and testing on the same structure always provide satisfactory accuracy, cross-testing on dissimilar structures leads to higher prediction errors, posing an extrapolation problem. This can be cured using heterogeneous training on databases that contain more than one structure, which results in a good trade-off between versatility and overall accuracy. Starting from a 3-body kernel trained this way, we build an efficient non-parametric 3-body force field that allows accurate prediction of structural properties at finite temperatures, following a newly developed scheme [A. Glielmo et al., Phys. Rev. B 95, 214302 (2017)]. We use this to assess the thermal stability of Ni19 nanoclusters at a fractional cost of full ab initio calculations. [ABSTRACT FROM AUTHOR]

Two different classes of molecular representations for use in machine learning of thermodynamic and electronic properties are studied. The representations are evaluated by monitoring the performance of linear and kernel ridge regression models on well-studied data sets of small organic molecules. One class of representations studied here counts the occurrence of bonding patterns in the molecule. These require only the connectivity of atoms in the molecule as may be obtained from a line diagram or a SMILES string. The second class utilizes the three-dimensional structure of the molecule. These include the Coulomb matrix and Bag of Bonds, which list the inter-atomic distances present in the molecule, and Encoded Bonds, which encode such lists into a feature vector whose length is independent of molecular size. Encoded Bonds’ features introduced here have the advantage of leading to models that may be trained on smaller molecules and then used successfully on larger molecules. A wide range of feature sets are constructed by selecting, at each rank, either a graph or geometry-based feature. Here, rank refers to the number of atoms involved in the feature, e.g., atom counts are rank 1, while Encoded Bonds are rank 2. For atomization energies in the QM7 data set, the best graph-based feature set gives a mean absolute error of 3.4 kcal/mol. Inclusion of 3D geometry substantially enhances the performance, with Encoded Bonds giving 2.4 kcal/mol, when used alone, and 1.19 kcal/mol, when combined with graph features. [ABSTRACT FROM AUTHOR]