Search

This research highlights the underestimated significance of cigarette paper as evidence at crime scenes. The primary objective is to distinguish cigarette paper from similar-looking alternatives, addressing the first research objective. The second objective involves identifying cigarette paper brands using attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy and machine learning (ML) algorithms. Accurate differentiation of cigarette paper from normal paper is emphasized. ATR-FTIR spectroscopy, coupled with principal component analysis (PCA) for dimensionality reduction, is employed for brand identification. Among fifteen ML algorithms compared, the CatBoost classifier excels for both objectives. This research presents a non-destructive, effective method for studying cigarette paper, contributing valuable insights to crime scene investigations. [Display omitted] • Forensic evaluation of cigarette paper utilizing ATR-FTIR spectroscopy and Machine learning algorithms. • Peak characterization and differentiation-distinguishing cigarette paper from other types. • Machine learning algorithm comparison: assessing discrimination across nine cigarette brands. • External validation of the dominant algorithm using unknown samples. [ABSTRACT FROM AUTHOR]

A critical aspect of software engineering is Software fault prediction which aims to identify and prevent errors in software systems before their release which can cause failures or issues for its users. Various techniques and tools have been developed to detect software faults, including static code analysis, dynamic testing, and machine learning-based approaches. In past few years, the world has seen a growing interest in the use of ML models for predicting software faults, as they can effectively analyse high dimensional datasets and detect complex patterns which are difficult for human experts to detect. However, developing accurate and reliable software fault detection models requires careful selection of data, feature engineering, and model evaluation. This purpose of this paper is to present a comprehensive analysis of potential applications and future research directions in the field of software fault detection. The study emphasizes the importance of identifying and addressing software faults to ensure the reliability and efficiency of software systems. Additionally, the paper outlines various approaches and techniques that can be employed for effective software fault detection. [ABSTRACT FROM AUTHOR]

One of the most dangerous threats in WSNs is botnet attacks, in which attackers use mutual communications between IoT devices to launch large-scale malicious activities. In this regard, developments in the field of effective and reliable means of defence against this type of threat, in particular, reliable methods for detecting, identifying, and countering botnet attacks, are becoming increasingly important and relevant. This paper presents a comprehensive study that applies machine learning techniques, namely Random Forest and XGBoost, to identify botnet attacks on IoT effectively. These algorithms are analyzed, compared, and shown to be highly effective in detecting complex patterns indicative of botnet activity, thus achieving a significant improvement in IoT security. The conducted research aims to make a useful contribution to the problem of securing WSNs and IoT in general. The results of the study demonstrated high accuracy in detecting attacks with an accuracy of 99.18% for XGBoost and Random Forest showed an accuracy of 99.21%. Thus, it was shown that the significance of applying machine learning techniques such as Random Forest and XGBoost can be one of the key approaches in combating botnet attacks and securing the IoT. The results of the work emphasize the promising application of machine learning techniques for effective defense against cyber threats and highlight the importance of further. [ABSTRACT FROM AUTHOR]

Techniques for classifying data using data mining are now a day prevalent in agriculture. The method of classifying seeds involves grouping various seed varieties according to their morphological characteristics. To accomplish categorization of the typical Charotar region (generally comprising Anand and Kheda districts of the Gujarat State of India) Gujarat Wheat (GW) varieties (TRITICUM – AESTIVUM) viz. GW 273, GW 496, GW 322, LOK-1, and GDW 1255 (TRITICUM – DURUM), Weka Explorer was used. The features used are area, perimeter, solidity, aspect ratio, major and minor axis of seed kernel, Hue, Saturation, Value, and SF1 (empirical). Features reduction was done using Information Gain (IG) and its modified version Gain Ratio (GR). This paper compares performance of Tree based data mining algorithms in classifying wheat varieties. For classification we used purely tree-based machine learning algorithms viz. J48, Random Forest, Hoeffding Tree, Logistic Model Tree (LMT), and REPTree. LMT- logistics regression method gives higher accuracy 96.4% compared to other classifiers. Hoeffding Tree classifiers stood second with 96% accuracy. For validation 10-fold cross validation was used. By reducing the number of folds in cross validation performance of most algorithms decreased except J48. The percentage of correctly classified instance increased for all algorithms when features were selected by GR except for J48. [ABSTRACT FROM AUTHOR]

Label distribution learning represents the relevance of labels to samples using description degree, which can provide richer semantic information, thus finding wider applications. Exploiting label correlations is an effective approach to narrow down the hypothesis space of label distribution learning models. In existing works that utilize low-rank assumptions or label linear dependence to mine correlations, it is assumed that a label can be linearly expressed by other labels. However, this assumption can only be satisfied when there are linear dependency relationships between labels, thus the label correlation obtained by such methods is subject to certain distortion. To address this issue, this paper assumes that labels can be linearly represented by the same set of bases. The correlation between labels is represented by sharing common bases. Specifically, the paper employs matrix factorization to extract bases that can be used to represent all labels. And then designs a label distribution learning algorithm based on the property of sharing the same set of bases of the ground truth label distribution and predict label distribution. The effectiveness of the algorithm is verified through experimental validation. Generally speaking, the algorithm presented in this paper achieves optimal performance at 73.15% of the cases, with the best average ranking. In the two-tailed t -test, the algorithm in this paper exhibits statistical superiority compared to all comparison algorithms. [ABSTRACT FROM AUTHOR]

Interactive teaching is very popular in the field of education, especially in college preschool education teaching. In this study, the experimental performance of the evaluation system based on scientific emotion construction is analyzed by the machine learning algorithm. The experimental results show that the classification model realized by the machine learning algorithm is feasible and effective for the construction of an interactive design teaching evaluation system of preschool education in colleges and universities. The main research content of this experiment is for the teaching evaluation text, this paper intends to use NB, KNN, LR, and SVM algorithms as the meta-learning algorithm in the framework, the classification effect of the meta-classifier is ranked as SVM > LR > KNN > NB. From the above data comparison, the SVM meta-classifier in the algorithm in this paper has achieved the best classification effect. The SVM meta-classifier based on the algorithm in this paper performs well, and the classification accuracy rate reaches 92.1%. This shows that this model is very suitable for studying the. And the classification accuracy rate reaches 92.1%. This shows that this model is very suitable for studying the construction of interactive teaching evaluation systems under emotional education. • The experimental results show that the classification model realized by machine learning algorithm. • This shows that this model is very suitable for studying the construction of interactive teaching. • The main research content of this experiment is for the teaching evaluation text. • This paper intends to use NB, KNN, LR and SVM algorithms as the meta-learning algorithm. [ABSTRACT FROM AUTHOR]

This paper considers the offline assignment of fixed priorities in partitioned preemptive real-time systems where tasks have precedence constraints. This problem is crucial in this type of systems, as having a good fixed priority assignment allows for an efficient use of the processing resources while meeting all the deadlines. In the literature, we can find several proposals to solve this problem, which offer varying trade-offs between the quality of their results and their computational complexities. In this paper , we propose a new approach, leveraging existing algorithms that are widely exploited in the field of Machine Learning: Gradient Descent, the Adam Optimizer, and Gradient Noise. We show how to adapt these algorithms to the problem of fixed priority assignment in conjunction with existing worst-case response time analyses. We demonstrate the performance of our proposal on synthetic task-sets with different sizes. This evaluation shows that our proposal is able to find more schedulable solutions than previous heuristics, approximating optimal but intractable algorithms such as MILP or brute-force, while requiring reasonable execution times. [ABSTRACT FROM AUTHOR]

Semi-Supervised Graph Clustering (SSGC) has emerged as a pivotal field at the intersection of graph clustering and semi-supervised learning (SSL), offering innovative solutions to intricate data analysis problems. However, despite its significance and wide-ranging applications, there exists a notable void in the literature—a comprehensive survey specifically dedicated to SSGC techniques and their diverse applications remains conspicuously absent. Addressing this gap, this paper presents a systematic and comprehensive review of SSGC methodologies, spanning from well-established approaches to cutting-edge developments. Through a meticulous categorization, critical examination, and insightful discussion of these techniques, this survey not only illuminates the current landscape of SSGC but also identifies unexplored avenues for exploration and innovation. In this paper we present a comprehensive survey of conventional graph construction, graph clustering, SSL methods, evaluation metrics and their primary development process. We then provide a taxonomy of SSGC techniques based on their structures and principles and thoroughly analyze and discuss the related models. Furthermore, we review the applications of SSGC in various fields. Lastly, we summarize the limitations of current SSGC algorithms and discuss the future directions of the field. [ABSTRACT FROM AUTHOR]

Advances over the past decade at the intersection of information fusion methods and Semi-Supervised Learning (SSL) are investigated in this paper that grapple with challenges related to limited labelled data. To do so, a bibliographic review of papers published since 2013 is presented, in which ensemble methods are combined with new machine learning algorithms. A total of 128 new proposals using SSL algorithms for ensemble construction are identified and classified. All the methods are categorised by approach, ensemble type, and base classifier. Experimental protocols, pre-processing, dataset usage, unlabelled ratios, and statistical tests are also assessed, underlining the major trends, and some shortcomings of particular studies. It is evident from this literature review that foundational algorithms such as self-training and co-training are influencing current developments, and that innovative ensemble techniques are continuing to emerge. Additionally, valuable guidelines are identified in the review for improving research into intrinsically semi-supervised and unsupervised pre-processing methods, especially for regression tasks. • A set of 128 recent semi-supervised ensemble methods is identified and analysed. • Semi-supervised approach, ensemble, base and preprocessing methods are identified. • The experimental protocol, settings, and data set properties are also categorised. • Some trends and shortcomings are identified in these methods and studies. [ABSTRACT FROM AUTHOR]

Achieving precision in the prediction of buildings' dynamic heat load is crucial for the advancement of smart heating systems. This research highlights the urgent need to enhance the accuracy of models predicting dynamic heat load. Through literature review, distinguished machine learning and regression algorithms were chosen to formulate prediction models. These models employ a data time-step adaptive strategy, a physics-guided loss function, and fundamental principles of heat transfer. Optimization algorithms of a mathematical nature were utilized to fine-tune the parameters and the framework of long short-term memory (LSTM) and multi-layer perceptron (MLP) models. An analytical comparison was undertaken between physics-guided models and those not guided by physics. Principal conclusions are: 1) Pelican optimization algorithm (POA)-LSTM model emerges as superior in heat load prediction accuracy of an office building, with percentage errors for actual and simulated datasets ranging from −6.7 % to 5.8 % and −5.2 %–4.5 %, respectively, and the mean absolute percentage error (MAPE) standing at 2.3 % and 1.3 %. 2) The linear regression model exhibits the lowest precision, with a MAPE of 17.5 % and 4.0 % for the 7-day prediction results in the actual and simulated datasets, respectively. These findings provide support for improving heat load prediction in heating systems. • The authors have implemented a total of 20 physically-guided models for predicting building thermal load, among which the POA-LSTM model demonstrates the highest accuracy. • The disparities in thermal load prediction accuracy between the physically-guided machine learning models and the physically-guided mathematical regression models have been identified. • The accuracy of thermal load prediction was compared between the five physical guidance models proposed in this paper and the five non-physical guidance models mentioned in other papers. [ABSTRACT FROM AUTHOR]

• Distinct DG Output Modes Validation: The simulation validates distinct DG output modes, impacting prediction. • Mode-Based State Estimation: The paper proposes an innovative mode-based state estimation using IPSO-DBSCAN clustering, overcoming k-means limitations. • Enhanced Data Recovery using DBSCAN: Enhanced data recovery employs DBSCAN with improved secondary clustering, guided by similarity and homogeneity, outperforming traditional methods. • BiGRU-PF State Estimation: BiGRU-PF enhances state estimation accuracy and robustness in distribution networks by detecting DG output modes, addressing k-means limitations, and improving data recovery. This paper proposes a new forecasting-aided state estimation (FASE) method for distribution systems that mitigates issues with uncertain distributed generation (DG) and lost measurements. We utilize an Improved Particle Swarm Optimization (IPSO)-optimized Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for examination of historical DG output data. Based on identified DG output modes, it facilitates precise state prediction and data reconstruction,. The proposed method employs a Bidirectional Gated Recurrent Unit (BiGRU) neural network for state prediction and a particle filter (PF) for final state filtering. The method verification is provided through Python simulations of the Distribution Transformer Unit (DTU)7k distribution network system, demonstrating improved accuracy and robustness against sudden load change and bad data in measurements. [ABSTRACT FROM AUTHOR]

In the last years, the massive deployment of new photovoltaic (PV) power plants has launched the connection of PV inverters to the electrical network. A single medium-sized ground-mounted PV plant may have thousands of these inverters linked to the grid and even more PV panels on the DC side. Upon reaching such a substantial magnitude of devices involved in grid-connected installations, the effective operation, management, predictive maintenance, and fault detection becomes increasingly challenging without integrating advanced prediction and automated anomaly detection systems. Artificial intelligence algorithms, grounded in data measurements, can be pivotal in addressing this challenge. This paper proposes several regression-based methods to predict PV plants' energy generation, which is useful for detecting transient and long-term anomalies. These models are trained using a Recursive Least Squares (RLS) method and require a minimum number of variables to yield satisfactory outcomes, which is one of the paper's contributions. They mainly rely on energy generation measurements and geolocation. Within the scope of this research, two distinct algorithms have been implemented and validated. The first algorithm, a simplified model, is engineered to analyse the daily efficiency variation, prioritizing the identification of faults and abnormal operational profiles in PV plants. On the other hand, the second algorithm adopts a more intricate model tailored to facilitate long-term diagnosis, enabling the assessment of PV efficiency degradation. In this work, both algorithms are described and their performance is validated using the historical data from more than 20 PV plants placed in different climatic regions. • Photovoltaic systems' modelling using linear regression models. • Photovoltaic fault detection using linear regression coefficients evaluation. • Automatic fault detection for photovoltaic plants with low sensors' availability. • Short-term and long-term efficiency trend evaluation of photovoltaic plants. [ABSTRACT FROM AUTHOR]

The emergence of IoT has introduced new security concerns, particularly the detection of large-scale network attacks initiated by compromised IoT devices. The diverse nature of IoT devices adds complexity to attack detection, necessitating the classification of devices based on their unique characteristics and behavior patterns. This paper presents a novel device classification method grounded in traffic differentials before and after IoT devices succumb to compromise. The study defines the concept of attack sensitivity and designs a method for calculating the value. The K-means algorithm is used to classify devices into predefined categories based on their attack sensitivity values. The effectiveness of this method is demonstrated through experiments on a public dataset. Additionally, a multi-class classification model is developed to identify newly added devices to the IoT environment, with good results in accuracy, precision, recall, F1 score, and FPR (0.999, 0.997–1,0.997–1,0.998–1,0–0.001). Overall, The device classification method proposed in this paper, based on attack sensitivity, not only enables a rational definition of device categories but also ensures precise classification for devices newly integrated into the IoT environment. This lays the foundation for achieving more precise attack detection and protection, advancing the field towards enhanced cybersecurity measures. [ABSTRACT FROM AUTHOR]

This comprehensive review focuses on the Autonomous Driving System (ADS), which aims to reduce human errors that are the reason for about 95% of car accidents. The ADS consists of six stages: sensors, perception, localization, assessment, path planning, and control. We explain the main state-of-the-art techniques used in each stage, analyzing 275 papers, with 162 specifically on path planning due to its complexity, NP-hard optimization nature, and pivotal role in ADS. This paper categorizes path planning techniques into three primary groups: traditional (graph-based, sampling-based, gradient-based, optimization-based, interpolation curve algorithms), machine and deep learning, and meta-heuristic optimization, detailing their advantages and drawbacks. Findings show that meta-heuristic optimization methods, representing 23% of our study, are preferred for being general problem solvers capable of handling complex problems. In addition, they have faster convergence and reduced risk of local minima. Machine and deep learning techniques, accounting for 25%, are favored for their learning capabilities and fast responses to known scenarios. The trend towards hybrid algorithms (27%) combines various methods, merging each algorithm's benefits and overcoming the other's drawbacks. Moreover, adaptive parameter tuning is crucial to enhance efficiency, applicability, and balancing the search capability. This review sheds light on the future of path planning in autonomous driving systems, helping to tackle current challenges and unlock the full capabilities of autonomous vehicles. [ABSTRACT FROM AUTHOR]

• A feature selection method based on binary dandelion algorithm is proposed. • The algorithm applies a sine-cosine operator and a restart strategy. • Mutual information and quick bit mutation improve the performance of the algorithm. • The algorithm performs well on datasets with larger dimensions. Feature selection (FS) is an important data preprocessing technology for machine learning and data mining. Metaheuristic algorithm (MH) has been widely used in feature selection because of its powerful search function. This paper presents an improved Binary Dandelion Algorithm using Sine Cosine operator and Restart strategy (SCRBDA) for feature selection. First, the sine cosine operator is used in the radius formula of the core dandelions (CD), which significantly enhances the ability of algorithm development and exploration. Secondly, the algorithm uses a restart strategy to increase its ability to get rid of local optimum. Thirdly, mutual information is used to guide the generation of some dandelions, which pays more attention to the correlation between the selected features and categories. Finally, quick bit mutation is used as the mutation strategy to improve the diversity of the population. The SCRBDA proposed in this paper was tested on 18 datasets of different sizes from UCI machine learning database. The SCRBDA was compared with 8 other classical feature selection algorithms, and the performance of the proposed algorithm was evaluated through feature subset size, classification accuracy, fitness value, and F1-score. The experimental results show that SCRBDA achieves the best performance, which has stronger feature reduction ability and achieves better overall performance on most datasets. Especially on large-scale datasets, SCRBDA can obtain extremely smaller feature subsets while maintaining much higher classification accuracy, and satisfactory F1-score. [ABSTRACT FROM AUTHOR]

In this paper, we apply the machine learning clustering algorithm Density Based Spatial Clustering of Applications with Noise (DBSCAN) to study the membership of stars in twelve open clusters (NGC 2264, NGC 2682, NGC 2244, NGC 3293, NGC 6913, NGC 7142, IC 1805, NGC 6231, NGC 2243, NGC 6451, NGC 6005 and NGC 6583) based on Gaia DR3 Data. This sample of clusters spans a variety of parameters like age, metallicity, distance, extinction and a wide parameter space in proper motions and parallaxes. We obtain reliable cluster members using DBSCAN as faint as G ∼ 20 mag and also in the outer regions of clusters. With our revised membership list, we plot color-magnitude diagrams and we obtain cluster parameters for our sample using ASteCA and compare it with the catalog values. We also validate our membership sample by spectroscopic data from APOGEE and GALAH for the available data. This paper demonstrates the effectiveness of DBSCAN in membership determination of clusters. [ABSTRACT FROM AUTHOR]

In the era of rapid paper publications in various venues, automatic citation recommendations would be highly useful to researchers when they write papers. Local citation recommendation aims to recommend possible papers to cite given local citation contexts. Previous work mainly computes the similarity score between citation contexts and cited papers on a one-to-one basis, which is quite time-consuming. We train a pair of neural network encoders that map citation contexts and all possible cited papers to the same vector space, respectively. After that, we index the positions of all cited papers in the vector space. This makes our process for searching recommended papers considerably faster. On the other hand, existing methods tend to recommend papers that are highly similar to each other, which makes recommendations lack diversity. Therefore, we extend our algorithm to perform multi-topic recommendations. We generate multi-topic training examples based on the index we mentioned earlier. Furthermore, we specially design a multi-group contrastive learning method to train our model so that it can distinguish different topics. Empirical experiments show that our model outperforms previous methods by a wide margin. Our model is also light weighted and has been deployed online so that researchers can use it to obtain recommended citations for their own paper in real-time. • Proposed FLCR algorithm with sentence-transformer & k-d tree. • Introduced multi-topic citation recommendation for diverse contexts. • Developed large-scale dataset with 1.7 million citation contexts for evaluation. • Demonstrated significant performance improvement over previous methods. • Deployed demo for real-time citation recommendations. [ABSTRACT FROM AUTHOR]

Intelligent Fault-Tolerant Control (IFTC) refers to the applications of machine learning algorithms for fault diagnosis and design of Fault-Tolerant Control (FTC). The overall goal of the FTC is to accommodate defects in the system components while they are in use and maintain stability with little to no performance reduction. These systems are crucial for mission-critical and safety-related applications where the safety of people is at stake and service continuity is crucial. In this review paper, a systematic study has been done for the development of FTC with machine learning, deep learning, and transfer learning algorithms. The challenges and limitations faced with their possible solutions through machine learning theories for the IFTC model are lined up. This paper guides researchers on the different possible types of machine learning algorithms and their advanced forms like deep learning and transfer learning. The differences among these are highlighted by the challenges and limitations of each. The paper is significant such that most of the important literature references from the Scopus database particularly related to important electrical and mechanical industrial problems have been discussed to guide the researchers who want to apply IFTC for specific industrial problems, being the research gap. Finally, future research directions for the development of IFTC are highlighted. [ABSTRACT FROM AUTHOR]

In this paper, to investigate the problem of radio wave propagation in complex environments, a new algorithm is proposed based on the equivalence principle algorithm and the domain decomposition method to rapidly and accurately predict electromagnetic scattering and radio wave propagation in half-space. First, equivalent surfaces are introduced to decompose the obstacle region and replace the scattering effect of obstacles within the equivalent surfaces with equivalent input and output electromagnetic currents. Then, the operator S, representing the relationship between the input and output equivalent electromagnetic currents on the equivalent surfaces, and the operator T, representing the interactions between the equivalent electromagnetic currents on each equivalent surface, are derived; this allows the calculation process of the overall spatial field in a complex environment to be simplified to the iterative calculation process of operators S and T. To reduce the complexity of calculating S, a neural network embedded with physical information is introduced to replace the calculation process. Finally, the training results are applied to an example and compared with the theoretical numerical results yielded by the method of moments in different cases with obstacles. The results reveal that the new method proposed in this paper is highly accurate and achieves a substantial improvement in calculation speed. Therefore, our proposed algorithm provides a new and reliable method for calculating the spatial field in complex environments. • The EPA is introduced to use equivalent surfaces to replace the scattering inside it. • The operators S and T are introduced into the process of solving the field. • A neural network with physical information embedded is applied to calculate the S. • Random undulating terrains are generated by Karhunen–Loève expansion. • MoM providing a good foundation for training is used to generate dataset. [ABSTRACT FROM AUTHOR]

Despite the existence of evidence-based rehabilitation strategies that address biomechanical deficits, the persistence of recurrent ankle problems in 70% of patients with acute ankle sprains highlights the unresolved nature of this issue. Artificial intelligence (AI) emerges as a promising tool to identify definitive predictors for ankle sprains. This paper aims to summarize the use of AI in investigating the ankle biomechanics of healthy and subjects with ankle sprains. Articles published between 2010 and 2023 were searched from five electronic databases. 59 papers were included for analysis with regards to: i). types of motion tested (functional vs. purposeful ankle movement); ii) types of biomechanical parameters measured (kinetic vs kinematic); iii) types of sensor systems used (lab-based vs field-based); and, iv) AI techniques used. Most studies (83.1%) examined biomechanics during functional motion. Single kinematic parameter, specifically ankle range of motion, could obtain accuracy up to 100% in identifying injury status. Wearable sensor exhibited high reliability for use in both laboratory and on-field/clinical settings. AI algorithms primarily utilized electromyography and joint angle information as input data. Support vector machine was the most used supervised learning algorithm (18.64%), while artificial neural network demonstrated the highest accuracy in eight studies. The potential for remote patient monitoring is evident with the adoption of field-based devices. Nevertheless, AI-based sensors are underutilized in detecting ankle motions at risk of sprain. We identify three key challenges: sensor designs, the controllability of AI models, and the integration of AI-sensor models, providing valuable insights for future research. [Display omitted] • Application of sensors and artificial intelligence to analyse ankle biomechanics. • Lab sensors predict parameters, paving the way for post-modern rehabilitation. • Field devices show potential for ankle injury prediction. • Focus on streamlined smart sensors and effective data management in the future. [ABSTRACT FROM AUTHOR]

Federated learning (FL) is a distributed machine learning method to protect users' privacy and security. It currently faces the following two problems (McMahan et al., 2017): (1) The performance of the global model degrades when dealing with Non-Independent Identically Distributed (Non-IID) data. Other existing classical methods do not have a rigorous theory based on error processing ; (2) In the process of global model aggregation, the classical FL algorithm either directly averages local models or solely considers the proportion of local data to assign weights to the models, without accounting for the discrepancies between the local models. In this paper, a federated aggregation algorithm Federated Maximum Mean Discrepancy (FedMMD) is proposed to address the deviation between the local models and Non-IID. First of all, this paper utilizes the Dilated Convolution Meet Transformer (DCMT) model for local model feature extraction. This approach aims to capture more feature information and minimize the impact of Non-IID scenarios. Secondly, the learning stability of the local model data participating in the aggregation is compared with Maximum Mean Discrepancy (MMD). The weights of the local models involved in the aggregation are determined using the SKNQ (Student–Keuls–Newman-Q) method and the entropy weight method. The SKNQ method calculates and compares the MMD across multiple clients, while the entropy weight method assigns different weights to clients based on their deviations. On the standard dataset, the FedAvg algorithm (McMahan et al., 2017), the FedProx algorithm (Li et al., 2020) and the FedMMD algorithm are compared. The experimental results demonstrate that the FedMMD algorithm, used in training the global model, not only enhances the accuracy of learning IID and Non-IID data but also improves the generalization capability of the global model. • An DCMT model is proposed to alleviate the model fluctuations caused by NonIID data. • An MMD method is proposed to compared horizontally before the model is aggregated. • Entropy weight method and the SKNQ method are combined to allocate local models' weight when they are fused into global model. [ABSTRACT FROM AUTHOR]

Accurate prediction of traffic flow is crucial to building a smart city. Given the nonlinearity of traffic flow, this paper proposes an extreme learning machine (ELM) algorithm based on residual correction and Tent chaos sequence combined with the DROP strategy. The algorithm is referred to as TRELM-DROP. The Tent chaos strategy and residual correction method reduce the impact of randomness in traffic flow. On this basis, the Tent and residual correction strategy avoids the weight optimization of the ELM algorithm using the iterative method. A DROP strategy is proposed in the proposed algorithm to improve its ability to predict traffic flow under varying conditions. A comprehensive comparison of 36 real-world datasets is presented in this paper, comparing TRELM-DROP with other benchmark models. The results show that the proposed algorithm can produce the best prediction performance regarding various prediction error metrics under various traffic conditions without iterative optimization. • An improved non-iterative traffic flow forecasting model is proposed. • Tent chaos strategy and residual correction strategy are developed. • A DROP strategy combining Dropout and Dropconnect is proposed. [ABSTRACT FROM AUTHOR]

In the real world, the sources of text often transcend the boundaries of generative and natural domains, which introduces challenges for existing linguistic steganalysis methods. The typical problem arises from the sample selection bias caused by training solely on a single-source domain, rendering the model incapable of inferencing across the entire generative-natural (GN) space. Moreover, there exists the problem of overlooking the sensitive discrepancies between generative text and natural text, which hampers model fitting. In this paper, we model steganalysis in a brand-new perspective by employing multi-task learning to build the main task and auxiliary tasks in the cross GN domain. The proposed Cross Generative-Natural Domain Multi-task Model (CG-NDMM) can concurrently address the two aforementioned issues through i) modeling steganalysis across the entire GN space, incorporating two auxiliary tasks alongside a main task, and ii) utilizing a feature representation transfer learning strategy to harmonize two sub-networks. Furthermore, we employ diverse steganography algorithms to construct the datasets, which comprise four types of texts (generative-cover, generative-steganographic, natural-cover, and natural-steganographic) derived from two public datasets, Movie and Twitter. The experiments on these datasets demonstrate the effectiveness of the proposed approach, showcasing its substantially superior performance over the comparative baseline methods. [Display omitted] • A cross generative-natural domain multi-task model is proposed for steganalysis. • Transfer learning captures features distinguishing generative text from natural ones. • Multi-task learning reduces sample bias, boosts cross-domain steganalysis performance. [ABSTRACT FROM AUTHOR]

Timely diagnosis and defect warning in power systems, related to the probability of a fault occurring, may be achieved by examining the correlations between various factors and evaluating the collected information. To check the performance of the machine learning models, a large amount of data is required in various operating conditions to feed as input to the algorithms. The primary focus of the conducted work is on the defect warning process, with additional studies discussing fault identification, type determination, and location. The stacking-based method is considered the main learning approach; however, the separate outcomes for the three single base methods, i.e., RF, SVM, and XGBoost, along with some other popular and strong algorithms have also been presented. In the warning study section, the proposed stacking approach achieved a minimum accuracy of 99.31 % while taking into consideration the noisy samples, and it has achieved a maximum accuracy of 100 % for all weather conditions with noise-free data. The fault identification component achieves an accuracy of 100 %, whereas the type determination part achieves an accuracy of 99.98 %. More details and results are provided in the paper. The modeling and simulations in this work were implemented in Python and MATLAB. • A defect warning procedure for transmission lines are introduced. • The effect of weather situations as external factors is regarded. • A stacking-based machine learning approach is proposed to detect and classify the defects. • A definite fault tracking process utilizing the stacking-based method is analyzed. • The accuracy of predictions has enhanced compared to individual models. [ABSTRACT FROM AUTHOR]

In this paper, a comparative study on the state of charge (SOC) estimation of the lithium-rich manganese-based battery (LRMB) has been conducted by systematically considering the equivalent circuit model (ECM), aging state and algorithms. The results show that the first-order RC model combined with the Extended Kalman filter (EKF) is more suitable for the SOC estimation of the LRMB when the battery decays less severely. The maximum value of root mean square error of SOC estimation error no exceeds 1.6 % under various dynamic operating conditions, which can well meet the practical application. Besides, the first-order model based EKF algorithm occupies least computing resource and simultaneously possesses the best convergence velocity against the initial SOC value deviation. When the LRMB suffers a severe aging, the SOC estimation performance decreases significantly with the max error exceeding 6 %, to overcome this, a new algorithm combined deep learning and Extend Kalman filter (EKF) is proposed, which exhibit high SOC estimation accuracy with the RMSE less than 0.16 %. In summary, a systematic study about the SOC estimation of LRMB has been carried out, which can provide great guidance for the future engineering application of LRMB. • Various Bayesian filtering algorithms and neural networks are systematically compared. • The aging states of LRMB are taken into account. • The different types of ECMs are adopted in the SOC estimation of LRMB. [ABSTRACT FROM AUTHOR]

In this paper, we focus on the two-block nonconvex and nonsmooth optimization with linear constraints, where the objective function is the sum of a convex but nonsmooth function and a smooth but nonconvex function. This problem encompasses many important applications in engineering and machine learning. Based on the splitting algorithm, making use of the quadratic approximation of the smooth part, and with the help of Armijo line search technique, we first propose a splitting augmented Lagrangian method with partial quadratic approximation. Under some mild conditions, the global convergence of the proposed method is proved. Moreover, when the gradient of the smooth part in the objective function is Lipschitz continuous, we then propose and analyze a variant of the aforementioned method without performing line search. We report some preliminary numerical results on solving nonconvex quadratic regularization problems to show the feasibility and effectiveness of the two proposed methods. Finally, we show applicability and encouraging efficiency of our methods by applying them to solve sparse signal recovery problems. [ABSTRACT FROM AUTHOR]

Printer source prediction is an important task when examining questioned documents. While some research has provided methods to predict the source printer of documents, with the advent of compatible consumables, printer prediction could become more complex and difficult. Predicting the source printer after replacing cartridges and identifying the source of printer cartridges are unresolved issues that are rarely addressed in current research. Herein, we introduce a novel technique to predict the manufacturer, model, and cartridges of laser printers (i.e., compatible, and original cartridges) used to produce a given document. Document samples produced using eight laser printers and 247 cartridges were collected to establish a dataset. Common manufacturers included HP, Canon, Lenovo, and Epson. After obtaining white-light images and three-dimensional profile images of printed characters, a morphological analysis was conducted by questioned document examiners (QDEs) using microscopy. Microscopic image features across a series of images were also extracted and analyzed using algorithms. Then, six high-dimensional reduction algorithms were used to obtain between- and within-printer variations as well as between- and within-cartridge variations. Finally, we conducted principal component analysis (PCA) and discriminant analysis. For 40 % of the samples, mixed discrimination analysis (MDA) and fixed discrimination analysis (FDA) were employed to predict the manufacturer, model and cartridge of laser printers used to produce the questioned printed document; the remaining 60 % samples comprised the training dataset. In the prediction of manufacturer, model and cartridge, our method achieved mean accuracies of 95.5 %, 97.5 %, and 90.2 %, respectively. Hence, this technique could reasonably aid in predicting the manufacturer, model, and cartridge of a laser printer, even if different cartridges are loaded into printers. [Display omitted] • A novel technique was provided to predict the manufacturer, model, and cartridges of laser printers. • Three-dimensional profiling combined with white-light microscopic images helps to improve the performance of the prediction. • This paper expands document examination from two dimensions to three dimensions. [ABSTRACT FROM AUTHOR]

Many power system operation and planning problems use the DC power flow approximation to address computational challenges from the nonlinearity of the AC power flow equations. The DC power flow simplifies the AC power flow equations to a linear form that relates active power flows to phase angle differences across branches, parameterized by coefficients based on the branches' susceptances. Inspired by techniques for training machine learning models, this paper proposes an algorithm that seeks optimal coefficient and bias parameters to improve the DC power flow approximation's accuracy. Specifically, the proposed algorithm selects the coefficient and bias parameter values that minimize the discrepancy, across a specified set of operational scenarios, between the power flows given by the DC approximation and the power flows from the AC equations. Gradient-based optimization methods like Broyden–Fletcher–Goldfarb–Shanno (BFGS), Limited-Memory BFGS (L-BFGS), and Truncated Newton Conjugate-Gradient (TNC) enable solution of the proposed algorithm for large systems. After an off-line training phase, the optimized parameters are used to improve the accuracy of the DC power flow during on-line computations. Numerical results show several orders of magnitude improvements in accuracy relative to a hot-start DC power flow approximation across a range of test cases. • Adaptive optimization for DC power flow coefficients b i j and bias parameters. • Scaling the algorithm with BFGS, L-BFGS, and TNC for large-scale power systems. • Superior accuracy of the algorithm shown by numerical results under various conditions. [ABSTRACT FROM AUTHOR]

This paper presents an algorithm for restoring AC power flow feasibility from solutions to simplified optimal power flow (OPF) problems, including convex relaxations, power flow approximations, and machine learning (ML) models. The proposed algorithm employs a state estimation-based post-processing technique in which voltage phasors, power injections, and line flows from solutions to relaxed, approximated, or ML-based OPF problems are treated similarly to noisy measurements in a state estimation algorithm. The algorithm leverages information from various quantities to obtain feasible voltage phasors and power injections that satisfy the AC power flow equations. Weight and bias parameters are computed offline using an adaptive stochastic gradient descent method. By automatically learning the trustworthiness of various outputs from simplified OPF problems, these parameters inform the online computations of the state estimation-based algorithm to both recover feasible solutions and characterize the performance of power flow approximations, relaxations, and ML models. Furthermore, the proposed algorithm can simultaneously utilize combined solutions from different relaxations, approximations, and ML models to enhance performance. Case studies demonstrate the effectiveness and scalability of the proposed algorithm, with solutions that are both AC power flow feasible and much closer to the true AC OPF solutions than alternative methods, often by several orders of magnitude in the squared two-norm loss function. • AC feasibility restoration algorithm for OPF relaxations, approximations, ML models. • Joint output from multiple OPFs for enhanced information utilization. • Adaptive stochastic gradient descent for optimal weight and bias. • Demonstrating algorithm effectiveness and scalability via numerical experiments. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is an efficacious tool that can be employed to enhance the pasteurization process through several means. This paper provides an analysis of the present advancements in ML-based pasteurization solutions and emphasizes the potential advantages that ML can offer to the pasteurization process in the dairy sector. ML can effectively monitor the pasteurization process in real-time, anticipate potential equipment malfunctions, enhance the efficiency of the process, and assess the quality of pasteurized products. The aforementioned skills have the potential to enhance the safety and quality of pasteurized products, concurrently leading to cost reduction and improved operational efficiency. Although there are various hurdles related to the implementation of ML. It is important to acknowledge the substantial potential benefits that ML may offer to the pasteurization process that will be beneficial for the dairy sector. With the advancement of ML models and the rising availability of data, it is anticipated that machine learning will assume a progressively significant role in the pasteurization process. This review provides detailed information to researchers, food and dairy technologists, and industrialists to explore new possibilities for applications of ML in the dairy pasteurization process. [ABSTRACT FROM AUTHOR]

This paper proposes a new, novel ranking method called Inverse-Reinforcement Learning Ranking. The main goal is to find a reward function representing the user's perceived utility after clicking on each result. It is necessary to utilize log information of all users' queries in the search engine dataset to reach this goal while assuming that the decisions (clicks) of the users are the best (optimal policy). The respective reward function is constructed using features extracted through a feature selection, and their corresponding weights are obtained by two optimization models, which are applicable for ranking results represented to the new users. In addition, new performance criteria were developed to illustrate the performance of the presented ranking method. To evaluate and test the proposed ranking algorithm, a real medium-sized dataset from a search engine is preprocessed and used in this research. Findings show promising results and decisive superiority over the default ranking method. It is illustrated that clicks on the top five results, top ten, and even the first results are remarkably improved by about 13–19% in all experiments, and the perplexity remarkably decreases by almost 23% after applying the ranking method. [ABSTRACT FROM AUTHOR]

[Display omitted] • A machine learning-assisted Impact Localization (IL) architecture is proposed. • Adaptive Sparse Noise Reduction Algorithm is studied to extract signal features. • RIME-optimized Dual-layer Support Vector Regression (RDSVR) is developed. • RDSVR enables to select optimal parameters and accurately predict impact location. In this paper, a machine-learning architecture with the integration of two strategies including data enhancement and adaptive generation scheme for Impact Localization (IL) are developed to address the aforementioned issues for location identification of impacts on composite laminates. Two main contributions are included in this research: First, response signals collected from low-speed impact experiments under various working conditions are denoised using Adaptive Sparse Noise Reduction Algorithm (ASNRA), which aims at maximizing the preservation of the original signal amplitude, thereby avoiding the underestimation of pulse features during denoising. Then a RIME-optimized Dual-layer Support Vector Regression (RDSVR) method for the real-time update of hyperparameters is implemented in the machine-learning architecture to realize IL. The superior performances of the IL architecture over different IL models are validated throughout the numerical examples in terms of stability and efficiency. Results demonstrate that proposed architecture has the ability to realize the accurate and robust IL of composite laminates. [ABSTRACT FROM AUTHOR]

Feature extraction techniques are widely used in fields such as machine learning, pattern recognition, and image processing. The quality of feature extraction is crucial to the generalization ability of a model. This paper proposes a feature extraction technique based on the Shapley value method and an improved Minimum Redundancy Maximum Relevance (mRMR) analysis method. The improved mRMR algorithm enhances the recognition ability for fitting effects of multi-variable feature subsets by traversing all possible combinations of input variable subsets. That is, the result of feature subset selection with fewer input variables does not affect the feature subset selection process when there are more input variables, thus avoiding the limitations of the original algorithm. The research results indicate that using the Shapley value method in conjunction with the improved mRMR algorithm proposed in this study can select the optimal feature subset containing fewer feature variables and achieve a lower MSE value. This contributes to achieving lower computational complexity and higher data fitting accuracy. This technique is applied to the feature extraction scenarios of power battery output power signals and instantaneous fuel consumption signals in hybrid electric vehicles, and constructs the optimal feature subsets for the aforementioned signals. • Exhibits superior global search capabilities, achieving improved data fitting results. • Improves the construction efficiency of the optimal feature subset. • Helps to improve the data fitting accuracy of the plant model. [ABSTRACT FROM AUTHOR]

In the field of machine learning, the issue of class imbalance is a common problem. It refers to an imbalance in the quantity of data collected, where one class has a significantly larger number of data compared to another class, which can negatively affect the classification efficiency of algorithms. Under-sampling methods address class imbalance by reducing the quantity of data in the majority class, thereby achieving a balanced dataset and mitigating the class imbalance problem. Traditional under-sampling methods based on k-means clustering either set the unified value of k (number of clusters) or determine it directly based on the quantity of data in the minority or majority class. This paper proposes an adaptive k-means clustering under-sampling algorithm that calculates an appropriate k for each dataset. After clustering the majority class dataset into k clusters, our algorithm calculates the distances between the data within each cluster and the cluster centroids from two perspectives and selects data based on these distances. Subsequently, the subset of the majority class dataset are combined with the minority class dataset to generate a new balanced dataset, which is then used for classification algorithms. The performance of our algorithm is evaluated on 45 datasets. Experimental results demonstrate that our algorithm can dynamically determine appropriate k for different datasets and output a balanced dataset, thus enhancing the classification efficiency of machine learning algorithms. This work can provide new algorithmic ensemble strategies for addressing class imbalance problem. [ABSTRACT FROM AUTHOR]

Due to the complex structure of the LFR fuel rod bundle, refined CFD simulation consumes significant computing resources and time, making it difficult to meet the immediate needs of engineering design and research. This paper introduces a method that integrates the sub-channel level thermal-hydraulic analysis code CorTAF-LBE with the BPNN algorithm to analyze the thermal-hydraulic phenomena within the LFR core. The method has been validated by refined simulation results of the THEADES test section, which indicate that the error is within an acceptable range. The feasibility of its application in LFR fuel assemblies was also evaluated. The flow characteristics and temperature distribution in unwrapped LFR rod bundles are investigated. Due to differences in fuel distribution and flow area, the temperature in the central channel is much higher than that in the side and trapezoidal channels. Three accident conditions, including inlet blockage, center blockage, and outlet blockage, were simulated. The corresponding peak temperatures of the fuel cladding were 670 K, 767 K, and 860 K, respectively, representing an increase of approximately 170 K compared to the nominal conditions. The analysis results effectively demonstrate the 3D temperature distribution of the fluid in the rod bundles, which is significant for reactor design and safety analysis of LFRs. [ABSTRACT FROM AUTHOR]

The inner states of ironmaking blast furnaces (BFs) govern their overall performance and thus are crucial for efficient and reliable BF production. However, the current control methods cannot directly consider the inner states because of the difficulty of accessing them. This paper introduces a machine learning (ML) model designed to predict the inner states according to injection parameters promptly. The model employs a modified ensemble learning method using data from a well-established mechanistic model. Two key modifications are implemented. A preprocessing method addresses the low prediction accuracy caused by large gradient data. A stack-based structure improves robustness across various inner states. Comparative analysis shows the proposed model predicts inner states with higher accuracy than existing ML models. Furthermore, the model outputs consistent resolutions while maintaining identical change trends for some key variables. The developed model offers a promising approach for implementing real-time BF prediction. [Display omitted] • A stacked-based model is developed for fast predicting blast furnaces (BFs) • It trains three individual learners and combines them to form a stronger predictor. • CFD data with the slip averaging algorithm applied is used to train the model. • The model's reliability in simulating BF performance is demonstrated. [ABSTRACT FROM AUTHOR]

In the last decade it became increasingly apparent the inability of technical metrics such as accuracy, sustainability, and non-regressiveness to well characterize the behavior of intelligent systems. In fact, they are nowadays requested to meet also ethical requirements such as explainability, fairness, robustness, and privacy increasing our trust in their use in the wild. Of course often technical and ethical metrics are in tension between each other but the final goal is to be able to develop a new generation of more responsible and trustworthy machine learning. In this paper, we focus our attention on machine learning algorithms and associated predictive models, questioning for the first time, from a theoretical perspective, if it is possible to simultaneously guarantee their performance in terms of both technical and ethical metrics towards machine learning algorithms that we can trust. In particular, we will investigate for the first time both theory and practice of deterministic and randomized algorithms and associated predictive models showing the advantages and disadvantages of the different approaches. For this purpose we will leverage the most recent advances coming from the statistical learning theory: Complexity-Based Methods, Distribution Stability, PAC-Bayes, and Differential Privacy. Results will show that it is possible to develop consistent algorithms which generate predictive models with guarantees on multiple trustworthiness metrics. • AI is nowadays requested to optimize both technical and ethical metrics. • Focus on ML algorithms and associated models (both deterministic and randomized). • We prove that it is possible to develop consistent algorithms and models. • We bound generalization in terms of both technical and ethical metrics. • We leverage the most recent advances coming from the statistical learning theory. [ABSTRACT FROM AUTHOR]

The removal of shadows from images is a classic problem in computer vision, aiming to restore the lighting in shadowed areas, thereby reducing the information interference and loss caused by the presence of shadows. In recent years, numerous excellent shadow removal algorithms have emerged, particularly with the rapid development of deep learning technology, which has disrupted traditional physics-based approaches and significantly improved the effectiveness of shadow removal. In this paper, we conduct a comprehensive survey of shadow removal methods published from 2017 to the present. We first introduce background knowledge about image shadow removal, providing detailed explanations of both physics-based and learning-based shadow removal methods. We analyze and compare these algorithms from both quantitative and qualitative perspectives, reassessing all models that provided open-source result sets according to uniform criteria. Additionally, we introduce commonly used datasets and evaluation metrics in the field. Finally, we discuss applications of shadow removal in specific scenarios, along with research challenges and opportunities in this domain. • Surveys taxonomy of recently image shadow removal techniques, datasets and performance metrics. • Systematic quantitative evaluations are summarized. • The impacts, applications and importance of shadow removal are explored. • Challenges and future direction are explored and analyzed. [ABSTRACT FROM AUTHOR]

Drowsiness detection is a test designed to detect a person's reaction ability and speed while in a state of fatigue. This type of drowsiness detection has many practical applications in areas like driver monitoring, transportation industry, and workplace safety. In this paper, a new machine learning drowsiness detection algorithm based on hybrid brain network feature extraction is proposed. First, the raw electroencephalogram signals are segmented into 30 s epochs, which are decomposed with wavelet packet transform into 7 layers to extract 30 s sleep characteristic subbands and the 7-th layer subbands. After that, 30 s subbands are divided into 5 s subbands with 50% overlap. At the same time, 30 s subbands are reconstructed by phase space reconstruction (PSR), whose parameters are optimized through energy valley optimizer. Then, based on mutual information and horizontal visual graph, a hybrid brain network is constructed with the 7-th layer subbands, 5 s subbands and PSR subbands, respectively. Finally, network features and time–frequency features are extracted and input into random forest classifier. The performance of the algorithm is evaluated through two tests: subject-independent test and subject-non-independent test. The results show that in Sleep-EDF dataset, accuracy, κ coefficient, recall, precision and F1-score obtained by the proposed algorithm can reach 94.19%, 88.39%, 95.37%, 92.53% and 93.93%, respectively. In Sleep-EDF Expanded dataset, accuracy, κ coefficient, recall, precision and F1-score can reach 93.65%, 87.65%, 94.53%, 92.11%, and 92.89%, respectively. [ABSTRACT FROM AUTHOR]

Interpretation of the prediction is vital for mission critical tasks. Accurate interpretation relies upon the generalization accuracy of the model. In this paper, we propose a modified gradient descent learning algorithm to improve the generalization ability of a Spiking Neural Network with time-varying weights (SNN-t). This algorithm is referred to as GradST, can help towards improving the interpretation of multiclass classification problems. We have transformed the SNN-t to a Generalized Additive Model (GAM) to provide interpretation. The resultant Spiking Additive Model (SAM) has the generalization ability of SNN-t and the interpretable characteristics of GAM for multiclass problems. We also propose a post-processing method to enable better visualization of multiple shape functions of GAMs, towards better relative interpretation for multiclass classification problems. The post-processing method utilizes the properties of multiclass GAMs to visually modify the shape functions to establish the importance of the feature in multiclass setting. We first evaluate the performance of SNN-t, trained with GradST and the SAM generated from it, on large public datasets. The SNN-t trained with GradST has better generalization accuracy than other SNN-t classifier and consequently, the SAM generated from it has better generalization accuracy than other state-of-the-art multiclass GAMs. Improved accuracy in SAM implies more reliable interpretation. Then, we evaluate the proposed post-processing method for multiclass GAMs to provide relative interpretation. It is observed that relative interpretation of multiclass GAM is more meaningful and reliable. • Modified gradient descent-based learning algorithm for a spiking neural classifier. • Transformation of spiking neural classifier to an interpretable classifier. • Improve generalization ability for reliable interpretation. • Post-processing method for relative interpretation. [ABSTRACT FROM AUTHOR]

The target net method has been the foundation of deep reinforcement learning since Deepmind first proposed it in 2015. Almost all the current popular reinforcement learning algorithms include target net. However, while the slowly updated target network improves the stability of the algorithm, it also reduces the performance of the algorithm. In this paper, the authors design a novel triple-network algorithm(TPN). TPN combines the temporal-difference(TD) algorithm and policy gradient(PG) theorem. Using three networks to estimate the state value(v), action value (q) , and policy(π). These networks have no primary or secondary distinction but are trained synchronously and influence each other. The author found that through this TPN architecture, the convergence and stability of the algorithm can be greatly improved without increasing the amount of calculation. Although it is only a basic framework at present. The calculation process of TPN is simple and easy to implement. Experiments prove that the convergence speed and stability of TPN in discrete cases are better than PPO. • Proposed a new deep reinforcement learning algorithm that does not rely on target networks. • Presented the complete derivation process of the TPN algorithm. • Verified the effectiveness of proposed algorithm through experiments. [ABSTRACT FROM AUTHOR]

The lighting requirements are subjective and one light setting cannot work for all. However, there is little work on developing smart lighting algorithms that can adapt to user preferences. To address this gap, this paper uses fuzzy logic and reinforcement learning to develop an adaptive lighting algorithm. In particular, we develop a baseline fuzzy inference system (FIS) using the domain knowledge, generating light recommendation based on a set of intuitive rules. These rules, derived from existing literature, are based on environmental conditions i.e. daily glare index, and user information including age, activity, and chronotype. Through a feedback mechanism, the user interacts with the algorithm, correcting the algorithm output to their preferences. We interpret these corrections as rewards to a Q-learning algorithm, which tunes the FIS parameters online to match the user preferences. The Q-learning is a model-free learning algorithm that learns to act optimally by interacting with the user and the rewards it receives. This allows the proposed algorithm to work in a model-free manner, effectively handling the uncertainties that might arise from the individualistic preferences of users. To the authors' best knowledge, this algorithm is pioneering work in designing intelligent algorithms for personalized lighting control, featuring several elements of novelty, including the number of environmental and user-related inputs, the continuous control of light intensity as opposed to common on/off control, and the ability to learn user preferences. The algorithm is implemented in a real aircraft cabin and is evaluated in an extensive user study. The implementation results demonstrate that the developed algorithm possesses the capability to learn user preferences while successfully adapting to a wide range of environmental conditions and user characteristics. This underscores its viability as a potent solution for intelligent light management, featuring advanced learning capabilities. • Intelligent lighting algorithm with the ability to learn from and adapt to user preferences. • A fuzzy inference system tunes lighting based on environment and user traits. • The developed algorithm is tested and verified via an in-depth user study. [ABSTRACT FROM AUTHOR]

Grid-interactive building control poses a critical challenge in the context of grid modernization and decarbonization. Recently, various artificial intelligence-based optimal control approaches have been proposed, providing innovative solutions to optimal building control problems. However, researchers and practitioners are now confronted with a dilemma when selecting appropriate control strategies, ranging from model-based (knowledge-incorporating) to learning-based (data-driven) approaches, and hybrid methods combining them. Although each algorithm in existing literature claims superiority over specific baselines, their performance has never been systematically compared and analyzed, owing to the absence of a unified platform for comprehensive evaluation. To fill this knowledge gap, we identify and implement all state-of-the-art approaches within a modular training and evaluation framework, assessing their efficacy in a grid-interactive building control problem. In this paper, we also introduce a streamlined hybrid method that complements existing hybrid approaches. Our comparative study reveals and quantifies the advantages of hybrid methods: on average, they achieve near-optimal control while requiring less than 14% of the online computation of traditional model-based methods. To achieve this performance, they need as few as 2% of the training samples of purely learning-based methods. Finally, we provide insights into the merits, limitations, and implementation of each method to help researchers better understand the state of the art and future directions. The software framework implemented in this study is open-sourced to facilitate future research. [ABSTRACT FROM AUTHOR]

Topological data analysis (TDA) is a powerful mathematical framework that extracts valuable insights about the shape and structure of complex datasets by identifying and analyzing underlying topological features, including connected components, holes, voids, and higher-dimensional counterparts. TDA achieves this by constructing a simplicial complex from the data, comprising simple shapes like points, lines, triangles, and higher-dimensional counterparts. The Vietoris–Rips complex (VRC), a widely used method, constructs simplicial complexes by measuring the distances between data points in a metric space. While the link between Intuitionistic Fuzzy Sets (IFSs) and TDA is clear through intuitionistic fuzzy distance measures, the literature has not explored the application of TDA for extracting topological features from uncertain and vague datasets using IFSs. In this paper, we propose a novel approach that leverages Intuitionistic Fuzzy Distance Measures (IFDMs) to generate the Intuitionistic Fuzzy Vietoris–Rips Complex (IFVRC) of IFSs. Specifically, we investigate the persistence of invariant relations of topological features between IFVRCs constructed using different distances, including Atanassov's Euclidean distance (l 2 I F S), Boran and Akay's distance (D), and Liu's distance measure (d L), over artificially generated IFSs. Furthermore, we demonstrate that the generation of IFVRC can be viewed as a Natural Hierarchical Clustering Algorithm (NHCA) for IFSs. Finally, we apply the proposed IFVRC as a Natural Hierarchical Clustering Algorithm to cluster intuitionistic fuzzy datasets related to Car and Building materials. Moreover, we incorporate a relatively large dataset, the GTZAN dataset from the Kaggle datasets repository, to classify music genres based on their topological information. • Topological analysis of intuitionistic fuzzy distance measures (IFDMs) have been conducted. • Utilizing IFDMs, Intuition Fuzzy Vietoris Rips Complex (IFVRC) generated. • It is shown how to extract topological features from IFVRC. • The persistence relations of topological features between IFVRCs have been investigated. • Finally, IFVRC is treated as a natural hierarchical clustering algorithm. [ABSTRACT FROM AUTHOR]

Bayesian network (BN) structure learning is the basis of BN applications and plays a pivotal role in many machine learning tasks. Whereas remarkable progress in structure learning has been achieved in the past, making further improvements in the efficiency and accuracy of structure learning is a significant challenge. In this paper, we propose an efficient skeleton learning approach-based hybrid algorithm (ESLH), which consists of two phases. In the constraint-based phase, a dynamic threshold (DTH) strategy and a skeleton learning method based on triangle breaking (SLTB) are proposed to learn the skeleton of a BN structure efficiently. DTH designs a dynamic threshold to remove redundant edges in the initial skeleton with little time overhead. By the result of DTH, SLTB first finds, tests and breaks triangles in the initial skeleton to efficiently remove redundant edges and then removes the remaining redundant edges to discover the final skeleton. In the score-and-search phase, ESLH employs the hill-climbing algorithm to find the highest-scored structure. We propose a novel strategy to divide this phase into three steps, both utilizing the learned skeleton to constrain the search space and preventing the errors of the learned skeleton from reducing the quality of the final learned structure. Extensive experiments on benchmark BNs validate the effectiveness of DTH and SLTB and demonstrate that ESLH is more than five times faster than the state-of-the-art structure learning algorithms while maintaining the highest average accuracy. [ABSTRACT FROM AUTHOR]

Machine learning and learning analytics are powerful tools that not only support researchers in the detailed measurement and enhancement of learning processes in various learning environments, but also enable the aggregation and synthesis of evidence regarding effective educational practices. This paper describes the development and application of machine learning algorithms aimed at semi-automatic selection of abstracts for a meta-analysis on the effects of simulation-based learning in higher education. The goal was to reduce the workload while also maintaining the transparency and objectivity of the selection process. The algorithms were trained, validated, and tested on a set of 3187 studies on simulation-based learning found in medical and educational databases collected before April 2018. Subsequently, they were utilized to classify abstracts for a follow-up meta-analysis consisting of 2373 studies (published between 2018 and 2020). The aim of training the algorithms was to predict studies' abstract eligibility based on words and combinations of words used in these abstracts. The application of the algorithms reduced the number of studies that had to be manually screened from 2373 to 711. A total of 458 studies from automatically selected abstracts were included in the full-text screening, indicating the high precision of the algorithms (also compared to the performance of human raters). We conclude that machine learning algorithms can be trained and used to classify abstracts for their eligibility, significantly reducing the workload for the researchers without diminishing objectivity and quality when updating systematic literature reviews with or without a meta-analysis. • Machine learning algorithms can be used to classify abstracts for their eligibility. • Machine learning can significantly reduce the workload without compromising quality, when updating meta-analysis. • Some machine learning algorithms can outperform human raters in precision. [ABSTRACT FROM AUTHOR]

Nowadays, the ASIC design is increasing in complexity, and PPA targets are pushed to the limit. The lack of physical information at the early design stages hinders precise timing predictions and may lead to design re-spins. In previous work, we successfully improved timing prediction at the post-placement stage using the Random Forest model, achieving 91.25% cell delay accuracy. Building upon this, we further investigate the potential of Ensemble Tree-based algorithms, specifically focusing on " Extremely Randomized Trees " and " Gradient Boosting ", to close the gap in cell delay accuracy. In this paper, we enrich the training dataset with new 16 nm industrial designs. The results demonstrate a substantial improvement, with an average cell delay accuracy of 92.01% and 84.26% on unseen data. The average Root-Mean-Square-Error is significantly reduced from 12.11 to 3.23 and 7.76 on unseen data. [ABSTRACT FROM AUTHOR]

Grassland restoration is a crucial method for preventing ecological degradation in grasslands. Unmanned Aerial Vehicles (UAVs) offer a promising solution to reduce extensive human labor and enhance restoration efficiency, given their fully automatic capabilities, yet their full potential remains exploited. This paper progresses this emerging technology for planning the grassland restoration. We undertake the first attempt to mathematically model the UAV-enabled restoration process as the maximization of restoration areas problem (MRAP). This model considers factors including limited UAV battery energy, grass seed weight, the number of restored areas, and their sizes. The MRAP is a composite problem involving trajectory design and area allocation, which are highly coupled and conflicting. Consequently, it requires solving two NP-hard subproblems: the variant Traveling Salesman Problem (TSP) and the Multidimensional Knapsack Problem (MKP) simultaneously. To address this complex problem, we introduce a novel cooperative memetic algorithm. The algorithm integrates an efficient heuristic algorithm, variant population-based incremental learning (PBIL), and a maximum-residual-energy-based local search (MRELS) strategy, referred to as CHAPBILM. The algorithm solves the two subproblems interlacedly by leveraging the interdependencies and inherent knowledge between them. The simulation results demonstrate that CHAPBILM successfully solves the MRAP on multiple instances in a near-optimal way. It also confirms the conflicts between trajectory design and area allocation. The effectiveness of CHAPBILM is further supported by comparisons with traditional optimization methods that do not exploit the interdependencies between the two subproblems. The proposed model and solution have the potential to be extended to other complex optimization problems in ecological protection and precision agriculture. • The maximization of restoration areas problem is first presented for the UAV-enabled grassland restoration method. • An energy-sensitive mathematical programming model is formulated for the maximization of restoration areas problem under the realistic constraints. • A novel cooperative memetic algorithm CHAPBILM is explored to effectively solve the maximization restoration areas problem, without ignoring the dependence between the two stages. • The simulation results demonstrate that CHAPBILM performs significantly better than the noncooperative optimization method for the problem, which also confirms the dependency relationship. [ABSTRACT FROM AUTHOR]

The Wasserstein distance in the optimal transport (OT) model is being widely used in machine learning applications due to its mathematical and geometric properties. Since the entropy regularized OT is a convex problem, and for its large-scale problems, many algorithms have been proposed. Existing methods mainly adopt iterative random projection techniques to approximate the Wasserstein distance in the OT model. In this paper, we employ the entropic domain decomposition algorithm with mini-batch learning to calculate the entropy regularized OT problem and investigate the stability of this algorithm. The algorithm performs on different cells of the domain instead of using the projection technique and it is based on a mini-batch loss function. The corresponding solution in each iteration is derived from the dual problem of entropy regularized OT on each decomposed cell. Once the solutions to the dual problem on all decomposed cells are obtained, the corresponding transport plan in each iteration can be obtained. We show that the transport plan computed by its dual variables is admissible, and we also give the sub-gradient approximation result. In addition, we exploit the statistical concentration result of the subsampled mini-batch Wasserstein distance on decomposed cells. Finally, the numerical experiments validate that the proposed algorithm can save computational time further. [ABSTRACT FROM AUTHOR]