Search

Conventional approaches for designing metasurfaces with the desired scattering response are often a time‐consuming and onerous process that heavily relies on the design experience and empirical methods. In this paper, a time and resource‐efficient design method is proposed for the synthesis of multiple metasurface polarizers with an efficient automated data generation process by utilizing the structure validation surrogate model (SVSM). Moreover, a novel attention and space‐to‐depth enhanced adaptive tandem network (ASE‐ATN) is proposed which integrates symmetry‐aware space‐to‐depth (SA‐SPD) along with the attention mechanism for efficient feature extraction. Several design examples of metasurfaces with linear polarization to orthogonal linear polarization (LP‐OLP) and/or linear polarization to circular polarization (LP‐CP) conversion capabilities are provided to verify the performance of the proposed method. The feasibility of the proposed method is demonstrated through the fabrication and measurement of a dual‐band dual‐polarization converter. [ABSTRACT FROM AUTHOR]

Accurate estimation of software effort and time in the software development process is a key activity to achieve the necessary product quality. However, underestimation or overestimation of effort has become a key challenge for software development. One of the main problems is the estimation with metrics from late stages, because the product must already be finished to make estimates. In this paper, the use of statistical models and machine learning approaches for software estimation are used in early stages such as software design, and a data set is presented with metric values of design artifacts with 37 software projects. As results, models for the estimation of development time and effort are proposed and validated through leave-one-out cross-validation. Further, machine learning techniques were employed in order to compare software projects estimations. Through the statistical tests, it was proven that the errors were not statistically different with the regression models for effort estimation. However, with Random Forest the best statistical results were obtained for estimating development time. [ABSTRACT FROM AUTHOR]

Background: Teachers often rely on the use of open‐ended questions to assess students' conceptual understanding of assigned content. Particularly in the context of mathematics; teachers use these types of questions to gain insight into the processes and strategies adopted by students in solving mathematical problems beyond what is possible through more close‐ended problem types. While these types of problems are valuable to teachers, the variation in student responses to these questions makes it difficult, and time‐consuming, to evaluate and provide directed feedback. It is a well‐studied concept that feedback, both in terms of a numeric score but more importantly in the form of teacher‐authored comments, can help guide students as to how to improve, leading to increased learning. It is for this reason that teachers need better support not only for assessing students' work but also in providing meaningful and directed feedback to students. Objectives: In this paper, we seek to develop, evaluate, and examine machine learning models that support automated open response assessment and feedback. Methods: We build upon the prior research in the automatic assessment of student responses to open‐ended problems and introduce a novel approach that leverages student log data combined with machine learning and natural language processing methods. Utilizing sentence‐level semantic representations of student responses to open‐ended questions, we propose a collaborative filtering‐based approach to both predict student scores as well as recommend appropriate feedback messages for teachers to send to their students. Results and Conclusion: We find that our method outperforms previously published benchmarks across three different metrics for the task of predicting student performance. Through an error analysis, we identify several areas where future works may be able to improve upon our approach. Lay Description: What is already known about this topic: Open‐ended questions are used by teachers in the domain of mathematics to assess their students understanding but automated support for these types of questions are limited in online learning platforms. Recent advancements in areas of machine learning and natural language processing have led to promising results in a range of domains and applications. What this paper adds: Emulating how teachers identify similar student answers can be used to build tools that support teachers in assessing and providing feedback to student open‐ended work. Implications for practice: Developing better automated supports can increase the amount of direct feedback students receive to guide their learning. [ABSTRACT FROM AUTHOR]

Nonconvex penalties are utilized for regularization in high-dimensional statistical learning algorithms primarily because they yield unbiased or nearly unbiased estimators for the parameters in the model. Nonconvex penalties existing in the literature such as SCAD, MCP, Laplace and arctan have a singularity at origin which makes them useful also for variable selection. However, in several high-dimensional frameworks such as deep learning, variable selection is less of a concern. In this paper, we present a nonconvex penalty which is smooth at origin. The paper includes asymptotic results for ordinary least squares estimators regularized with the new penalty function, showing asymptotic bias that vanishes exponentially fast. We also conducted simulations to better understand the finite sample properties and conducted an empirical study employing deep neural network architecture on three datasets and convolutional neural network on four datasets. The empirical study based on artificial neural networks showed better performance for the new regularization approach in five out of the seven datasets. [ABSTRACT FROM AUTHOR]

As the cradle of scientific research institutions and the cultivation of innovative talents, high-level universities are of certain theoretical and practical significance to investigate the impact of high-level universities on the innovation of local enterprises under the realistic background of the revival of machine learning. This paper takes A-share listed enterprises in Shanghai and Shenzhen stock markets from 2013 to 2019 as samples, and empirically examines the relationship between high-level universities and local enterprise innovation from the perspective of "double first-class" universities. The study found that high-level universities have a significant role in promoting the innovation of local enterprises, and a series of robustness tests have also confirmed the conclusions of this paper. Mechanism analysis shows that high-level universities promote local enterprise innovation mainly by promoting human capital output and school-enterprise cooperation. Heterogeneity analysis shows that executive gender, government intervention, and differences in regional development levels have significant heterogeneity effects. This paper has obvious policy implications: China should guide high-level universities to cultivate high-quality talents, cultivate positive school-enterprise cooperation, promote the sharing of innovative resources, and promote the high-quality development of the real economy. [ABSTRACT FROM AUTHOR]

Aim: Leaf traits are central to plant function, and key variables in ecosystem models. However recently published global trait maps, made by applying statistical or machine‐learning techniques to large compilations of trait and environmental data, differ substantially from one another. This paper aims to demonstrate the potential of an alternative approach, based on eco‐evolutionary optimality theory, to yield predictions of spatio‐temporal patterns in leaf traits that can be independently evaluated. Innovation: Global patterns of community‐mean specific leaf area (SLA) and photosynthetic capacity (Vcmax) are predicted from climate via existing optimality models. Then leaf nitrogen per unit area (Narea) and mass (Nmass) are inferred using their (previously derived) empirical relationships to SLA and Vcmax. Trait data are thus reserved for testing model predictions across sites. Temporal trends can also be predicted, as consequences of environmental change, and compared to those inferred from leaf‐level measurements and/or remote‐sensing methods, which are an increasingly important source of information on spatio‐temporal variation in plant traits. Main conclusions: Model predictions evaluated against site‐mean trait data from > 2,000 sites in the Plant Trait database yielded R2 = 73% for SLA, 38% for Nmass and 28% for Narea. Declining species‐level Nmass, and increasing community‐level SLA, have both been recently reported and were both correctly predicted. Leaf‐trait mapping via optimality theory holds promise for macroecological applications, including an improved understanding of community leaf‐trait responses to environmental change. [ABSTRACT FROM AUTHOR]

Nonparametric and machine learning methods are flexible methods for obtaining accurate predictions. Nowadays, data sets with a large number of predictors and complex structures are fairly common. In the presence of item nonresponse, nonparametric and machine learning procedures may thus provide a useful alternative to traditional imputation procedures for deriving a set of imputed values used next for the estimation of study parameters defined as solution of population estimating equation. In this paper, we conduct an extensive empirical investigation that compares a number of imputation procedures in terms of bias and efficiency in a wide variety of settings, including high-dimensional data sets. The results suggest that a number of machine learning procedures perform very well in terms of bias and efficiency. [ABSTRACT FROM AUTHOR]

Python is a popular programming language, and a large part of its appeal comes from diverse libraries and extension modules. In the bloom of data science and machine learning, Python frontend with C/C++ native implementation achieves both productivity and performance and has almost become the standard structure for many mainstream software systems. However, feature discrepancies between two languages such as exception handling, memory management, and type system can pose many safety hazards in the interface layer using the Python/C API. In this paper, we carry out an empirical study of the Python/C API on evolution and bug patterns. The evolution analysis includes Python/C API design in CPython compilers and its usage in mainstream software. By designing and applying a static analysis toolset, we reveal the evolution and usage statistics of the Python/C API and provide a summary and classification of 9 common bug patterns. In Pillow, a widely used Python imaging library, we find 48 bugs, 19 of which are undiscovered before. Our toolset can be easily extended to access different types of syntactic bug‐finding checkers, and our systematical taxonomy to classify bugs can guide the construction of more highly automated and high‐precision bug‐finding tools. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a neurodegenerative condition that progresses over decades. Early detection of individuals at high risk of future progression toward AD is likely to be of critical significance for the successful treatment and/or prevention of this devastating disease. In this paper, we present an empirical study to characterize how predictable an individual subjects' future AD trajectory is, several years in advance, based on rich multi-modal data, and using modern deep learning methods. Crucially, the machine learning strategy we propose can handle different future time horizons and can be trained with heterogeneous data that exhibit missingness and non-uniform follow-up visit times. Our experiments demonstrate that our strategy yields predictions that are more accurate than a model trained on a single time horizon (e.g. 3 years), which is common practice in prior literature. We also provide a comparison between linear and nonlinear models, verifying the well-established insight that the latter can offer a boost in performance. Our results also confirm that predicting future decline for cognitively normal (CN) individuals is more challenging than for individuals with mild cognitive impairment (MCI). Intriguingly, however, we discover that prediction accuracy decreases with increasing time horizon for CN subjects, but the trend is in the opposite direction for MCI subjects. Additionally, we quantify the contribution of different data types in prediction, which yields novel insights into the utility of different biomarkers. We find that molecular biomarkers are not as helpful for CN individuals as they are for MCI individuals, whereas magnetic resonance imaging biomarkers (hippocampus volume, specifically) offer a significant boost in prediction accuracy for CN individuals. Finally, we show how our model's prediction reveals the evolution of individual-level progression risk over a five-year time horizon. Our code is available at https://github.com/batuhankmkaraman/mlbasedad. [ABSTRACT FROM AUTHOR]

As discarding superfluous instances in data sets shortens the learning process, it also increases learning performance because of eliminating noisy data. Instance selection methods are commonly utilized to undertake the abovementioned tasks. In this paper, we propose a new supervised instance selection algorithm called Border Instances Reduction using Classes Handily (BIRCH). BIRCH considers k-nearest neighbors of each instance and selects instances that have neighbors from the only same class, namely, but not having neighbors from the different classes. It has been compared with one traditional and four state-of-the-art instance selection algorithms by using fifteen data sets from various domains. The empirical results show BIRCH well delivers the trade-off between accuracy rate and reduction rate by tuning the number of neighbors. Furthermore, the proposed method guarantees to yield a high classification accuracy. The source code of the proposed algorithm can be found in https://github.com/fatihaydin1/BIRCH. [ABSTRACT FROM AUTHOR]

In continual learning over deep neural networks (DNNs), the rehearsal strategy , in which the previous exemplars are jointly trained with new samples, is commonly employed for the purpose of addressing catastrophic forgetting. Unfortunately, due to the memory limit, rehearsal-based techniques inevitably cause the class imbalance issue leading to a DNN biased toward new tasks having more samples. Existing works mostly focus on correcting such a bias in the fully connected layer or classifier. In this paper, we newly discover that class imbalance tends to make old classes even more highly correlated with their similar new classes in the feature space, which turns out to be the major reason behind catastrophic forgetting, called inter-task forgetting. To alleviate inter-task forgetting, we propose a novel class incremental learning method, called attractive & repulsive training (ART) , which effectively captures the previous feature space into a set of class-wise flags , and thereby makes old and new similar classes less correlated in the new feature space. In our empirical study, our ART method is observed to be quite effective to improve the performance of the state-of-the-art methods by substantially mitigating inter-task forgetting. Our implementation is available at: https://github.com/bigdata-inha/ART/. [ABSTRACT FROM AUTHOR]

Predictive Process Monitoring (PPM) has been integrated into process mining use cases as a value-adding task. PPM provides useful predictions on the future of the running business processes with respect to different perspectives, such as the upcoming activities to be executed next, the final execution outcome, and performance indicators. In the context of PPM, Machine Learning (ML) techniques are widely employed. In order to gain trust of stakeholders regarding the reliability of PPM predictions, eXplainable Artificial Intelligence (XAI) methods have been increasingly used to compensate for the lack of transparency of most of predictive models. Multiple XAI methods exist providing explanations for almost all types of ML models. However, for the same data, as well as, under the same preprocessing settings or same ML models, generated explanations often vary significantly. Corresponding variations might jeopardize the consistency and robustness of the explanations and, subsequently, the utility of the corresponding model and pipeline settings. This paper introduces a framework that enables the analysis of the impact PPM-related settings and ML-model-related choices may have on the characteristics and expressiveness of the generated explanations. Our framework provides a means to examine explanations generated either for the whole reasoning process of an ML model, or for the predictions made on the future of a certain business process instance. Using well-defined experiments with different settings, we uncover how choices made through a PPM workflow affect and can be reflected through explanations. This framework further provides the means to compare how different characteristics of explainability methods can shape the resulting explanations and reflect on the underlying model reasoning process. [ABSTRACT FROM AUTHOR]

The uncertainty associated with option price predictions has largely been overlooked in the literature. This paper aims to fill this gap by quantifying such uncertainty using conformal prediction. Conformal prediction is a model-agnostic procedure that constructs prediction intervals, ensuring valid coverage in finite samples without relying on distributional assumptions. Through the simulation of synthetic option prices, we find that conformal prediction generates prediction intervals for gradient boosting machines with an empirical coverage close to the nominal level. Conversely, non-conformal prediction intervals exhibit empirical coverage levels that fall short of the nominal target. In other words, they fail to contain the actual option price more frequently than expected for a given coverage level. As anticipated, we also observe a decrease in the width of prediction intervals as the size of the training data increases. However, we uncover significant variations in the width of these intervals across different options. Specifically, out-of-the-money options and those with a short time-to-maturity exhibit relatively wider prediction intervals. Then, we perform an empirical study using American call and put options on individual stocks. We find that the empirical results replicate those obtained in the simulation experiment. • Conformal prediction quantifies well the uncertainty of option price predictions. • Conformal prediction intervals have empirical coverage near the nominal level. • Non-conformal prediction intervals have empirical coverage below the nominal level. • Large variations in prediction intervals are found for American call and put options. • Larger intervals are found for out-of-the-money and short time-to-maturity options. [ABSTRACT FROM AUTHOR]

This paper presents a plantar pressure sensor system (P2S2) integrated in the insoles of shoes to detect thirteen commonly used human movements including walking, stooping left and right, pulling a cart backward, squatting, descending, ascending stairs, running, and falling (front, back, right, left). Six force sensitive resistors (FSR) sensors were positioned on critical pressure points on the insoles to capture the electrical signature of pressure change in the various movements. A total of 34 adult participants were tested with the P2S2. The pressure data were collected and processed using a Principal Component Analysis (PCA) for input to the multiple machine learning (ML) algorithms, including k-NN, neural network and Support-Vector Machine (SVM) algorithms. The ML models were trained using four-fold cross-validation. Each fold kept subject data independent from other folds. The model proved effective with an accuracy of 86%, showing a promising result in predicting human movements using the P2S2 integrated in shoes. [ABSTRACT FROM AUTHOR]

• QSPR models for predicting critical temperature of binary mixtures are built. • Linear and non-linear methods are adopted for QSPR modeling. • Internal and external validations and application domain are conducted. • The R ext 2 and AARD of the best QSPR model are 0.98 and 1.27%, respectively. The critical temperature is an important parameter in the design and selection of binary organic mixtures. Rapid and accurate prediction has been a focus of research. Strong nonlinear relationships exist between molecular characteristics and critical temperature. Developing nonlinear models is an important measure to improve the prediction accuracy. In this paper, 56 different QSPR models are developed to predict the critical temperature using 14 types of mixture descriptors and four modeling methods (MLR, XGBoost, RBFNN and SVM). A dataset containing 2540 data points is adopted. Results show x 1 d 1 + x 2 d 2 is the optimum mixture descriptor type, and the accuracy of the nonlinear models is better than that of the linear models. The model combining x 1 d 1 + x 2 d 2 and SVM exhibits the best predictive power; the R ext 2 , RMSE ext , and AARD are 0.988, 10.057 K, and 1.27%, respectively. For this model, 78.98% of the data have an absolute deviation of less than 5 K, and the accuracy is better than that of existing QSPR models for critical temperature of mixtures. The application domain analysis shows the model has good performance for novel binary organic mixtures. In addition, compared with the empirical calculation methods for predicting the critical temperature, the results show the developed model has higher reliability. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the impact of pattern leakage during data preprocessing on the reliability of Machine Learning (ML) based intrusion detection systems (IDS). Data leakage, also known as pattern leakage, occurs during data preprocessing when information from the testing set is used in training, leading to overfitting and inflated accuracy scores. Our study uses three well-known intrusion detection datasets: NSL-KDD, UNSW-NB15, and KDDCUP99. We preprocess the data to create versions with and without pattern leakage and train and test six ML models: Decision Tree (DT), Gradient Boosting (GB), K-neighbours (KNN), Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR). Our results show that building IDS models with data leakage leads to higher accuracy but is unreliable. Additionally, we find that some algorithms are more sensitive to data leakage than others, as seen by the drop in model accuracy when built without leakage. To address this problem, we provide suggestions for mitigating data leakage in the training process and analyzing the sensitivity of different algorithms. Overall, our study emphasizes the importance of addressing data leakage in the training process to ensure the reliability of ML-based IDS models. [ABSTRACT FROM AUTHOR]

Improving the accuracy of coating lifetime prediction on small sample data has been an urgent issue to be addressed. In this paper, based on in-situ electrochemical impedance spectroscopy data of multilayer Cr/GLC coatings, a lifetime prediction formula related to the coating failure mechanism is developed, which provides a quantitative basis for estimating the coating lifetime. Furthermore, the combination of the mechanistic prediction model and the ANN + RF integrated machine learning model can further increase the model's prediction accuracy, reach 97.9%, and provide a new method for predicting coating performance and lifetime in deep-sea environments. [Display omitted] • Developed a model for predicting the lifetime of multilayer structural coatings applicable in deep-sea environments. • Prediction of coating lifetime by an empirical function relationship. • Six different machine learning models were screened to determine the superiority of the RF model. • The integrated ANN + RF model shows higher accuracy than the RF model for corrosion prediction. • Prediction accuracy of 97.9% for integrated ANN + RF + mechanistic models modified by mechanistic empirical models. [ABSTRACT FROM AUTHOR]

Asphalt pavement compaction is a process of rearranging asphalt particles to reduce voids and increase density. Current empirical methods can sometimes cause compaction problems, particularly when new materials are implemented. This paper develops an innovative monitoring system to determine the compaction condition of asphalt pavement based on particle kinematics. Eleven asphalt mixtures were compacted in the laboratory, and two field compaction projects were carried out. A particle-size wireless sensor, SmartRock, was used to collect the particle kinematic behaviors during compaction. A compaction classification model and a density prediction model were established to predict the compaction condition. Similar particle rotation characteristics were identified under gyratory compaction and roller compaction, allowing the predictive models developed in the laboratory to estimate the field compaction. The predictions are in agreement with the measurement densities, which verifies the applicability of using the particle kinematics and intelligent model to predict the compaction of asphalt pavement. • An innovative methodology is developed to sense particle kinematics and determine the compactability of asphalt pavements. • Three-stage compaction patterns are identified based on particle rotation during compaction. • The field compaction condition can be predicted using the prediction model trained by the laboratory data. • Particle rotation, compaction energy, and material properties are critical parameters for compaction prediction. • Both qualitative and quantitative models are developed for asphalt pavement compaction prediction. [ABSTRACT FROM AUTHOR]

• Reviewed traditional & adaptive empirical methods of pavement performance modeling. • Discussed pros & cons of traditional and adaptive performance modeling methods. • Identified the gaps in empirical modeling methods and provided future study directions. This paper conducts a comprehensive review of empirical methods of pavement performance modeling. The paper firstly analyzes performance measures used in existing pavement performance modeling. Then, based on whether pavement performance models can be automatically updated with newly observed data, this paper classifies the empirical methods into traditional modeling methods and adaptive modeling methods and discusses their main features, strengths, and weaknesses. It is found that, in the traditional modeling methods, comprehensive modeling methods typically can well incorporate both temporal and spatial characteristics of pavement performance but the interpretation capability of performance models based on machine learning techniques needs to be enhanced; the adaptive modeling methods describe the deterioration behavior of pavement performance in a more accurate way but need to improve the setting of updating conditions and the evaluation of updating effects. Finally, recommendations are made for future studies. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new mathematical model of language interactions considering bilingualism. We assume diffusive and convective language spreads with language exchange terms. The resulting model consists of a coupled system of partial differential equations for three functions. We develop a finite element approximation of the mathematical model using implicit and partially explicit time schemes. In addition, we present a partial learning approach for this problem using a partially explicit discretization. In this method, we train a Deep Neural Network to predict the values of the difficult-to-compute (implicit) part of the solution at some observed points. We then perform linear temporal interpolation and spatial interpolation using the Proper Orthogonal Decomposition (POD) and the Discrete Empirical Interpolation Method (DEIM). To test the proposed approach and the model, we consider two model problems. Each of them simulates different language situations. For each problem, we compute the relative errors of the proposed approach and define the different parts of the error. The numerical results show that the proposed approach can provide good accuracy while reducing computational costs. [ABSTRACT FROM AUTHOR]

In this paper, we present a rigorous comparative study to assess and identify the most effective machine learning model for blast loading prediction. Blast loads are known to produce catastrophic effects including structural collapse and personnel fatality. Accurate and efficient prediction of these extreme loads using empirical methods and numerical solvers remains a challenging problem. Machine learning provides a promising alternative solution, which has been increasingly used in various engineering applications. However, there is seldom any analysis or justification of the selection of machine learning method that would lead to the best performance for such applications. For example, most existing machine learning-based approaches for blast loading prediction utilise the classic multi-layer perceptron (MLP) network with no justifications of their suitability and efficiency nor attempts of leveraging other state-of-the-art neural network architectures. In this study, four well-known machine learning models, including one ensemble tree method and three neural networks of different types, are investigated to demonstrate the effectiveness of different machine learning methods for blast loading prediction. It is showcased using BLEVE (boiling liquid expanding vapour explosion) pressure prediction that the Transformer model achieves the best performance, reaching a relative error of 3.5% and R 2 0.997 that outperforms the existing MLP approach (relative error 6.0%, R 2 0.985) with a clear margin. This study shows that the Transformer network is an effective tool for prediction of blast loading from BLEVE as well as other explosion sources. • Conducted a comparative study of state-of-the-art machine learning models for blast loading prediction. • Rigorously evaluated GBDT, MLP, ResNet, and Transformer models for BLEVE pressure prediction. • Demonstrated that Transformer produces the most accurate prediction, significantly outperforming the widely adopted MLP. • Applied the developed Transformer model to BLEVE experimental data, achieving a 25.4% relative error. [ABSTRACT FROM AUTHOR]

The majority of the intrusion detection solutions proposed using machine learning and deep learning approaches are based on known attack classes only. Comprehensive threat detection systems should consider both known and unknown attacks. Rapidly changing network environment and the advanced tools and techniques used by adversaries to launch new sophisticated attacks highlight a growing need to build intrusion detection systems that are more realistic, diverse, and robust to detect known and unknown attacks. We employed deep-learning models in our experiments to detect unknown threats, never introduced before to the model. This paper also studied the bias issues in connection with unknown threats detection. Many recent research studies based on conventional machine learning may report biased results and restricted training due to relying only on a single dataset; thus, there are existing threats that the model is unaware of, although the model may have high accuracy (in the known territories). This study presents a realistic IDS approach in which a deep learning classifiers' ensemble is trained on four benchmark IDS datasets for testing the unknown attack instances. Specifically, the model has no prior knowledge of some labels and traffic patterns in those experiments. The architecture proposed builds a deep learning ensemble using classifiers well-known to process and produce good results for sequential data. Our empirical results indicate that the proposed ensemble model can detect a range of unknown attacks with reasonable performance measures and a practical approach towards building a comprehensive IDS solution. [ABSTRACT FROM AUTHOR]