Search

Gas is the transaction-fee metering system of the Ethereum network. Users of the network are required to select a gas price for submission with their transaction, creating a risk of overpaying or delayed/unprocessed transactions involved in this selection. In this work, we investigate data in the aftermath of the London Hard Fork and shed insight into the transaction dynamics of the network after this major fork. As such, this paper provides an update on work previous to 2019 on the link between EthUSD/BitUSD and gas price. For forecasting, we compare a novel combination of machine learning methods such as Direct-Recursive Hybrid LSTM, CNN-LSTM, and Attention-LSTM. These are combined with wavelet threshold denoising and matrix profile data processing toward the forecasting of block minimum gas price, on a 5-min timescale, over multiple lookaheads. As the first application of the matrix profile being applied to gas price data and forecasting that we are aware of, this study demonstrates that matrix profile data can enhance attention-based models; however, given the hardware constraints, hybrid models outperformed attention and CNN-LSTM models. The wavelet coherence of inputs demonstrates correlation in multiple variables on a 1-day timescale, which is a deviation of base free from gas price. A Direct-Recursive Hybrid LSTM strategy is found to outperform other models, with an average RMSE of 26.08 and R2 of 0.54 over a 50-min lookahead window compared to an RMSE of 26.78 and R2 of 0.452 in the best-performing attention model. Hybrid models are shown to have favorable performance up to a 20-min lookahead with performance being comparable to attention models when forecasting 25–50-min ahead. Forecasts over a range of lookaheads allow users to make an informed decision on gas price selection and the optimal window to submit their transaction in without fear of their transaction being rejected. This, in turn, gives more detailed insight into gas price dynamics than existing recommenders, oracles and forecasting approaches, which provide simple heuristics or limited lookahead horizons. [ABSTRACT FROM AUTHOR]

Machine Learning is an interesting tool for stance recognition in a large-scale context, in terms of data size, but also regarding the topics and themes addressed or the languages employed by the participants. Public consultations of citizens using online participatory democracy platforms offer this kind of setting and are good use cases for automatic stance recognition systems. In this paper, we propose to use three datasets of public consultations, in order to train a model able to classify the stance of a citizen within a text, towards a proposal or a debate question. We studied stance detection in several contexts: using data from an online platform without interactions between users, using multilingual data from online debates that are in one language, and using data from online intra-multilingual debates, which can contain several languages inside the same unique debate discussion. We propose several baselines and methods in order to take advantage of the different available data, by comparing the results of models using out-of-dataset annotations, and binary or ternary annotations from the target dataset. We finally proposed a self-supervised learning method to take advantage of unlabelled data. We annotated both the datasets with ternary stance labels and made them available. [ABSTRACT FROM AUTHOR]

Traffic accidents directly influence public safety and economic development; thus, the prevention of traffic accidents is of great importance in urban transportation. The accurate prediction of traffic accidents can assist traffic departments to better control and prevent accidents. Thus, this paper proposes a deep learning method named attention-based residual dilated network (ARDN), to extract essential information from multi-source datasets and enhance accident prediction accuracy. The method utilizes bidirectional long short-term memory to model sequential information and incorporates an attention mechanism to recalibrate weights. Furthermore, a dilated residual layer is adopted to capture long term information effectively. Feature encoding is also employed to incorporate natural language descriptions and point-of-interest data. Experimental evaluations of datasets collected from Austin and Houston demonstrate that ARDN outperforms a range of machine learning methods, such as logistic regression, gradient boosting, Xgboost, and deep learning methods. The ablation experiments further confirm the indispensability of each component in the proposed method. [ABSTRACT FROM AUTHOR]

A pharmaceutical supply chain (PSC) is a system of processes, operations, and organisations for drug delivery. This paper provides a new PSC mathematical cost model, which includes Blockchain technology (BT), that can improve the safety, performance, and transparency of medical information sharing in a healthcare system. We aim to estimate the costs of the BT-based PSC model, select algorithms with minimum prediction errors, and determine the cost components of the model. After the data generation, we applied four Supervised Learning algorithms (k-nearest neighbour, decision tree, support vector machine, and naive Bayes) combined with two Evolutionary Computation algorithms (ant colony optimization and the firefly algorithm). We also used the Feature Weighting approach to assign appropriate weights to all cost model components, revealing their importance. Four performance metrics were used to evaluate the cost model, and the total ranking score (TRS) was used to determine the most reliable predictive algorithms. Our findings show that the ACO-NB and FA-NB algorithms perform better than the other six algorithms in estimating the costs of the model with lower errors, whereas ACO-DT and FA-DT show the worst performance. The findings also indicate that the shortage cost, holding cost, and expired medication cost more strongly influence the cost model than other cost components. [ABSTRACT FROM AUTHOR]

Data envelopment analysis (DEA) has been widely applied to evaluate the performance of banks, enterprises, governments, research institutions, hospitals, and other fields as a non-parametric estimation method for evaluating the relative effectiveness of research objects. However, the composition of its effective frontier surface is based on the input-output data of existing decision units, which makes it challenging to apply the method to predict the future performance level of other decision units. In this paper, the Slack Based Measure (SBM) model in DEA method is used to measure the relative efficiency values of decision units, and then, eleven machine learning models are used to train the absolute efficient frontier to be applied to the performance prediction of new decisions units. To further improve the prediction effect of the models, this paper proposes a training set under the DEA classification method, starting from the training-set sample selection and input feature indicators. In this paper, regression prediction of test set performance based on the training set under different classification combinations is performed, and the prediction effects of proportional relative indicators and absolute number indicators as machine-learning input features are explored. The robustness of the effective frontier surface under the integrated model is verified. An integrated models of DEA and machine learning with better prediction effects is proposed, taking China's regional carbon-dioxide emission (carbon emission) performance prediction as an example. The novelty of this work is mainly as follows: firstly, the integrated model can achieve performance prediction by constructing an effective frontier surface, and the empirical results show that this is a feasible methodological technique. Secondly, two schemes to improve the prediction effectiveness of integrated models are discussed in terms of training set partitioning and feature selection, and the effectiveness of the schemes is demonstrated by using carbon-emission performance prediction as an example. This study has some application value and is a complement to the existing literature. [ABSTRACT FROM AUTHOR]

Fast and accurate smoke detection is very important for reducing fire damage. Due to the complexity and changeable nature of smoke scenes, existing smoke detection technology has the problems of a low detection rate and a high false negative rate, and the robustness and generalization ability of the algorithms are not high. Therefore, this paper proposes a smoke detection model based on the improved YOLOv5. First, a large number of real smoke and synthetic smoke images were collected to form a dataset. Different loss functions (GIoU, DIoU, CIoU) were used on three different models of YOLOv5 (YOLOv5s, YOLOv5m, YOLOv5l), and YOLOv5m was used as the baseline model. Then, because of the problem of small numbers of smoke training samples, the mosaic enhancement method was used to randomly crop, scale and arrange nine images to form new images. To solve the problem of inaccurate anchor box prior information in YOLOv5, a dynamic anchor box mechanism is proposed. An anchor box was generated for the training dataset through the k-means++ clustering algorithm. The dynamic anchor box module was added to the model, and the size and position of the anchor box were dynamically updated in the network training process. Aiming at the problem of unbalanced feature maps in different scales of YOLOv5, an attention mechanism is proposed to improve the network detection performance by adding channel attention and spatial attention to the original network structure. Compared with the traditional deep learning algorithm, the detection performance of the improved algorithm in this paper was is 4.4% higher than the mAP of the baseline model, and the detection speed reached 85 FPS, which is obviously better and can meet engineering application requirements. [ABSTRACT FROM AUTHOR]

Cyber-attacks and unauthorized application usage have increased due to the extensive use of Internet services and applications over computer networks, posing a threat to the service's availability and consumers' privacy. A network Intrusion Detection System (IDS) aims to detect aberrant traffic behavior that firewalls cannot detect. In IDSs, dimension reduction using the feature selection strategy has been shown to be more efficient. By reducing the data dimension and eliminating irrelevant and noisy data, several bio-inspired algorithms have been employed to improve the performance of an IDS. This paper discusses a modified bio-inspired algorithm, which is the Grey Wolf Optimization algorithm (GWO), that enhances the efficacy of the IDS in detecting both normal and anomalous traffic in the network. The main improvements cover the smart initialization phase that combines the filter and wrapper approaches to ensure that the informative features will be included in early iterations. In addition, we adopted a high-speed classification method, the Extreme Learning Machine (ELM), and used the modified GWO to tune the ELM's parameters. The proposed technique was tested against various meta-heuristic algorithms using the UNSWNB-15 dataset. Because the generic attack is the most common attack type in the dataset, the primary goal of this paper was to detect generic attacks in network traffic. The proposed model outperformed other methods in minimizing the crossover error rate and false positive rate to less than 30%. Furthermore, it obtained the best results with 81%, 78%, and 84% for the accuracy, F1-score, and G-mean measures, respectively. [ABSTRACT FROM AUTHOR]

Wind power has become an indispensable part of renewable energy development in various countries. Due to the high cost and complex structure of wind turbines, it is important to design a method that can quickly and effectively determine the structural health of the generator set. This research proposes a method that could determine structural damage or weaknesses in the blades at an early stage via a model to monitor the sound of the wind turbine blades, so as to reduce the quantity of labor required and frequency of regular maintenance, and to repair the damage rapidly in the future. This study used the operating sounds of normal and abnormal blades as a dataset. The model used discrete wavelet transform (DWT) to decompose the sound into different frequency components, performed feature extraction in a statistical measure, and combined with outlier exposure technique to train a deep neural network model that could capture abnormal values deviating from the normal samples. In addition, this paper observed that the performance of the monitoring model on the MIMII dataset was also better than the anomaly detection models proposed by other papers. [ABSTRACT FROM AUTHOR]

Artificial intelligence, particularly machine learning, has begun to permeate various real-world applications and is continually being explored in automatic music generation. The approaches to music generation can be broadly divided into two categories: rule-based and data-driven methods. Rule-based approaches rely on substantial prior knowledge and may struggle to handle large datasets, whereas data-driven approaches can solve these problems and have become increasingly popular. However, data-driven approaches still face challenges such as the difficulty of considering long-distance dependencies when handling discrete-sequence data and convergence during model training. Although the diffusion model has been introduced as a generative model to solve the convergence problem in generative adversarial networks, it has not yet been applied to discrete-sequence data. This paper proposes a transformer-based diffusion model known as MelodyDiffusion to handle discrete musical data and realize chord-conditioned melody generation. MelodyDiffusion replaces the U-nets used in traditional diffusion models with transformers to consider the long-distance dependencies using attention and parallel mechanisms. Moreover, a transformer-based encoder is designed to extract contextual information from chords as a condition to guide melody generation. MelodyDiffusion can automatically generate diverse melodies based on the provided chords in practical applications. The evaluation experiments, in which Hits@k was used as a metric to evaluate the restored melodies, demonstrate that the large-scale version of MelodyDiffusion achieves an accuracy of 72.41% (k = 1). [ABSTRACT FROM AUTHOR]

Machine learning and deep learning have made tremendous progress over the last decade and have become the de facto standard across a wide range of image, video, text, and sound processing domains, from object recognition to image generation. Papers invited for this Special Issue considered aspects of this problem, including: Machine learning for optimization problems; Statistical learning; End-to-end machine learning; Graph neural networks; Combining classic optimization algorithms and machine learning; Mathematical models of problems for machine learning; Optimization method for machine learning; Evolutionary computation and optimization problems; Applications such as scheduling problems, smart cities, etc. The articles presented in this Special Issue provide insights into fields related to "Advances in Machine Learning and Mathematical Modeling for Optimization Problems", including models, performance evaluation and improvements, and application developments. These methods also sometimes include classic search and optimization algorithms for machine learning, such as Monte Carlo Tree Search in AlphaGO. [Extracted from the article]

The protection of critical infrastructure such as water treatment and water distribution systems is crucial for a functioning economy. The use of cyber-physical systems in these systems presents numerous vulnerabilities to attackers. To enhance security, intrusion detection systems play a crucial role in limiting damage from successful attacks. Machine learning can enhance security by analysing data patterns, but several attributes of the data can negatively impact the performance of the machine learning model. Data in critical water system infrastructure can be difficult to work with due to their complexity, variability, irregularities, and sensitivity. The data involve various measurements and can vary over time due to changes in environmental conditions and operational changes. Irregular patterns and small changes can have significant impacts on analysis and decision making, requiring effective data preprocessing techniques to handle the complexities and ensure accurate analysis. This paper explores data preprocessing techniques using a water treatment system dataset as a case study and provides preprocessing techniques specific to processing data in industrial control to yield a more informative dataset. The results showed significant improvement in accuracy, F1 score, and time to detection when using the preprocessed dataset. [ABSTRACT FROM AUTHOR]

The field of statistical computing is rapidly developing and evolving. Shifting away from the formerly siloed landscape of mathematics, statistics, and computer science, recent advancements in statistical computing are largely characterized by a fusing of these worlds; namely, programming, software development, and applied statistics are merging in new and exciting ways. There are numerous drivers behind this advancement, including open movement (encompassing development, science, and access), the advent of data science as a field, and collaborative problem-solving, as well as practice-altering advances in subfields such as artificial intelligence, machine learning, and Bayesian estimation. In this paper, we trace this shift in how modern statistical computing is performed, and that which has recently emerged from it. This discussion points to a future of boundless potential for the field. [ABSTRACT FROM AUTHOR]

Protein-protein docking reveals the process and product in protein interactions. Typically, a protein docking works with a docking model sampling, and then an evaluation method is used to rank the near-native models out from a large pool of generated decoys. In practice, the evaluation stage is the bottleneck to perform accurate protein docking. In this paper, PointNet, a deep learning algorithm based on point cloud, is applied to evaluate protein docking models. The proposed architecture is able to directly learn deep representations carrying the geometrical properties and atomic attributes from the 3D structural data of protein decoys. The experimental results show that the informative representations can benefit our proposed method to outperform other algorithms. [ABSTRACT FROM AUTHOR]

The classical numerical methods for differential equations are a well-studied field. Nevertheless, these numerical methods are limited in their scope to certain classes of equations. Modern machine learning applications, such as equation discovery, may benefit from having the solution to the discovered equations. The solution to an arbitrary equation typically requires either an expert system that chooses the proper method for a given equation, or a method with a wide range of equation types. Machine learning methods may provide the needed versatility. This article presents a method that uses an optimization algorithm for a parameterized approximation to find a solution to a given problem. We take an agnostic approach without dividing equations by their type or boundary conditions, which allows for fewer restrictions on the algorithm. The results may not be as precise as those of an expert; however, our method enables automated solutions for a wide range of equations without the algorithm's parameters changing. In this paper, we provide examples of the Legendre equation, Painlevé transcendents, wave equation, heat equation, and Korteweg–de Vries equation, which are solved in a unified manner without significant changes to the algorithm's parameters. [ABSTRACT FROM AUTHOR]

With the development of artificial intelligence, techniques such as machine learning, object detection, and trajectory tracking have been applied to various traffic fields to detect accidents and analyze their causes. However, detecting traffic accidents using closed-circuit television (CCTV) as an emerging subject in machine learning remains challenging because of complex traffic environments and limited vision. Traditional research has limitations in deducing the trajectories of accident-related objects and extracting the spatiotemporal relationships among objects. This paper proposes a traffic accident detection method that helps to determine whether each frame shows accidents by generating and considering object trajectories using influence maps and a convolutional neural network (CNN). The influence maps with spatiotemporal relationships were enhanced to improve the detection of traffic accidents. A CNN is utilized to extract latent representations from the influence maps produced by object trajectories. Car Accident Detection and Prediction (CADP) was utilized in the experiments to train our model, which achieved a traffic accident detection accuracy of approximately 95%. Thus, the proposed method attained remarkable results in terms of performance improvement compared to methods that only rely on CNN-based detection. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel creation method of feature graphics by deep learning algorithms based on a channel attention module consisting of a separable deep convolutional neural network and an SENet network. The main innovation of this method is that the image feature of sample images is extracted by convolution operation and the key point matrix is obtained by channel weighting calculation to create feature graphics within the channel attention module. The main problem of existing image generation methods is that the complex network training and calculation process affects the accuracy and efficiency of image generation. It greatly reduced the complexity of image generation and improved the efficiency when we trained the image generation network with the feature graphic maps. To verify the superiority of this method, we conducted a comparative experiment with the existing method. Additionally, we explored the influence on the accuracy and efficiency of image generation of the channel number of the weighting matrix based on the test experiment. The experimental results demonstrate that this method highlights the image features of geometric lines, simplifies the complexity of image generation and improves the efficiency. Based on this method, images with more prominent line features are generated from the description text and dynamic graphics are created for the display of the images generated, which can be applied in the construction of smart museums. [ABSTRACT FROM AUTHOR]

The potential of machine learning (ML) tools for enhancing geotechnical analysis has been recognized by several researchers. However, obtaining a sufficiently large digital dataset is a major technical challenge. This paper investigates the use of transfer learning, a powerful ML technique, used for overcoming dataset size limitations. The study examines two scenarios where transfer learning is applied to tunnel support analysis. The first scenario investigates transferring knowledge between a ground formation that has been well-studied to a new formation with very limited data. The second scenario is intended to investigate whether transferring knowledge is possible from a dataset that relies on simplified tunnel support analysis to a more complex and realistic analysis. The technical process for transfer learning involves training an Artificial Neural Network (ANN) on a large dataset and adding an extra layer to the model. The added layer is then trained on smaller datasets to fine-tune the model. The study demonstrates the effectiveness of transfer learning for both scenarios. On this basis, it is argued that, with further development and refinement, transfer learning could become a valuable tool for ML-related geotechnical applications. [ABSTRACT FROM AUTHOR]

Extractive Question Answering, also known as machine reading comprehension, can be used to evaluate how well a computer comprehends human language. It is a valuable topic with many applications, such as in chatbots and personal assistants. End-to-end neural-network-based models have achieved remarkable performance on these tasks. The most frequently used approach to extract answers with neural networks is to predict the answer's start and end positions in the document, independently or jointly. In this paper, we propose another approach that considers all words in an answer jointly. We introduce an encoder-decoder model to learn from all words in the answer. This differs from previous works. which usually focused on the start and end and ignored the words in the middle. To help the encoder-decoder model to perform this task better, we employ evaluation-based reinforcement learning with different reward functions. The results of an experiment on the SQuAD dataset show that the proposed method can outperform the baseline in terms of F1 scores, offering another potential approach to solve the extractive QA task. [ABSTRACT FROM AUTHOR]

Neurological diseases are a significant health threat, often presenting through abnormalities in electroencephalogram (EEG) signals during seizures. In recent years, machine learning (ML) technologies have been explored as a means of automated EEG pathology diagnosis. However, existing ML-based EEG binary classification methods largely focus on extracting EEG-related features, which may lead to poor performance in classifying EEG signals by overlooking potentially redundant information. In this paper, we propose a novel Kruskal–Wallis (KW) test-based framework for EEG pathology detection. Our framework first divides EEG data into frequency sub-bands using wavelet packet decomposition and then extracts statistical characteristics from each selected coefficient. Next, the piecewise aggregation approximation technique is used to obtain the aggregated feature vectors, followed by the KW statistical test methodology to select significant features. Finally, three ensemble learning classifiers, random forest, categorical boosting (CatBoost), and light gradient boosting machine, are used to classify the extracted significant features into normal or abnormal classes. Our proposed framework achieves an accuracy of 89.13%, F1-score of 87.60%, and G-mean of 88.60%, respectively, outperforming other competing techniques on the same dataset, which shows the great promise in EEG pathology detection. [ABSTRACT FROM AUTHOR]

Modern approaches to computing consumer price indices include the use of various data sources, such as web-scraped data or scanner data, which are very large in volume and need special processing techniques. In this paper, we address one of the main problems in the consumer price index calculation, namely the product classification, which cannot be performed manually when using large data sources. Therefore, we conducted an experiment on automatic product classification according to an international classification scheme. We combined 9 different word-embedding techniques with 13 classification methods with the aim of identifying the best combination in terms of the quality of the resultant classification. Because the dataset used in this experiment was significantly imbalanced, we compared these methods not only using the accuracy, F1-score, and AUC, but also using a weighted F1-score that better reflected the overall classification quality. Our experiment showed that logistic regression, support vector machines, and random forests, combined with the FastText skip-gram embedding technique provided the best classification results, with superior values in performance metrics, as compared to other similar studies. An execution time analysis showed that, among the three mentioned methods, logistic regression was the fastest while the random forest recorded a longer execution time. We also provided per-class performance metrics and formulated an error analysis that enabled us to identify methods that could be excluded from the range of choices because they provided less reliable classifications for our purposes. [ABSTRACT FROM AUTHOR]

The formal concept of lattice plays a vital role in knowledge discovery. Reduction of the attribute has many applications in machine learning technology and data mining fields. In this paper, we introduce an object ranking concept to define a consistency set and the reduction of the attributes by structural features. An incomplete information system works on the three-way concepts using the SE-ISI Context. The granular was emphasized with join (meet) irreducible sets using the object ranking concepts. A dual operator is defined based on the object ranking concepts and its properties and conditions are verified. Hence, this elaborates on the four kinds of reduction of the attributes. The ordered pairs give the knowledge of the attributes that deal with the interval set of both the approximation of rough set theory concerning the objects. Therefore, the relationship between four kinds of reduction of the attribute was appropriate to access the consistency set using the object ranking concepts by some of the theorems and examples. [ABSTRACT FROM AUTHOR]

Robots are now widely used in product disassembly lines, which significantly improves end-of-life (EOL) product disassembly efficiency. Most of the current research on disassembly line balancing problems focuses on decomposing one product. More than one product can be disassembled concurrently, which can further improve the efficiency. Moreover, uncertainty such as the depreciation of EOL products, may result in disassembly failure. In this research, a stochastic multi-product robotic disassembly line balancing model is established using an AND/OR graph. It takes the precedence relationship, cycle constraint, and disassembly failure into consideration to maximize the profit and minimize the energy consumption for disassembling multiple products. A Pareto-improved multi-objective brainstorming optimization algorithm combined with stochastic simulation is proposed to solve the problem. Furthermore, by conducting experiments on some real cases and comparing with four state-of-the-art multi-objective optimization algorithms, i.e., the multi-objective discrete gray wolf optimizer, artificial bee colony algorithm, nondominated sorting genetic algorithm II, and multi-objective evolutionary algorithm based on decomposition, this paper validates its excellent performance in solving the concerned problem. [ABSTRACT FROM AUTHOR]

In this paper, a machine-learning model is utilized to estimate the temperature-dependent moduli of neat, thermally reduced graphene and covalently functionalized graphene/epoxy nanocomposites. In addition, the governed mathematical expressions have been used to solve the buckling problem of beams fabricated from such nanocomposites in the presence of a thermal gradient. In order to do so, an energy-based method including the shear deformable beam hypothesis is used. The beam structure is rested on the Winkler–Pasternak substrate. The reported verifications demonstrate the impressive precision of the presented ML model, as well as the buckling response of the under-study structures. Finally, in the framework of some numerical case studies, the impact of several parameters on the buckling of nanocomposite beams is depicted. The results of this study delineate that temperature has a vital role in the determination of the critical buckling load that the nanocomposite structures can endure. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is a well-known and reliable technology that enables a machine to simulate human behavior. While the major theme of AI is to make a smart computer system that thinks like a human to solve awkward problems, machine learning allows a machine to automatically learn from past information without the need for explicit programming. In this analysis, we aim to derive the idea of Aczel–Alsina aggregation operators based on an intuitionistic fuzzy soft set. The initial stage was the discovery of the primary and critical Aczel–Alsina operational laws for intuitionistic fuzzy soft sets. Subsequently, we pioneer a range of applicable theories (set out below) and identify their essential characteristics and key results: intuitionistic fuzzy soft Aczel–Alsina weighted averaging; intuitionistic fuzzy soft Aczel–Alsina ordered weighted averaging; intuitionistic fuzzy soft Aczel–Alsina weighted geometric operators; and intuitionistic fuzzy soft Aczel–Alsina ordered weighted geometric operators. Additionally, by utilizing certain key information, including intuitionistic fuzzy soft Aczel–Alsina weighted averaging and intuitionistic fuzzy soft Aczel–Alsina weighted geometric operators, we also introduce the theory of the weighted aggregates sum product assessment method for intuitionistic fuzzy soft information. This paper also introduces a multi-attribute decision-making method, which is based on derived operators for intuitionistic fuzzy soft numbers and seeks to assess specific industrial problems using artificial intelligence or machine learning. Finally, to underline the value and reasonableness of the information described herein, we compare our obtained results with some pre-existing information in the field. This comparison is supported by a range of numerical examples to demonstrate the practicality of the invented theory. [ABSTRACT FROM AUTHOR]

The analysis of epilepsy electro-encephalography (EEG) signals is of great significance for the diagnosis of epilepsy, which is one of the common neurological diseases of all age groups. With the developments of machine learning, many data-driven models have achieved great performance in EEG signals classification. However, it is difficult to select appropriate hyperparameters for the models to file a specific task. In this paper, an evolutionary algorithm enhanced model is proposed, which optimizes the fixed weights of the reservoir layer of the echo state network (ESN) according to the specific task. As evaluating a feature extractor relies heavily on the classifiers, a new feature distribution evaluation function (FDEF) using the label information of EEG signals is defined as the fitness function, which is an objective way to evaluate the performance of a feature extractor that not only focuses on the degree of dispersion, but also considers the relation amongst triplets. The performance of the proposed method is verified on the Bonn University dataset with an accuracy of 98.16% and on the CHB-MIT dataset with the highest sensitivity of 96.14%. The proposed method outperforms the previous EEG methods, as it can automatically optimize the hyperparameters of ESN to adjust the structure and initial parameters for a specific classification task. Furthermore, the optimization direction by using FDEF as the fitness of MFO no longer relies on the performance of the classifier but on the relative separability amongst classes. [ABSTRACT FROM AUTHOR]

To prevent the proliferation of space debris and stabilize the space environment, active debris removal (ADR) has increasingly gained public concern. Considering the complexity of space operations and the viability of ADR missions, it would be necessary to schedule the ADR process in order to remove as much debris as possible. This paper presents an active debris removal mission planning problem, devoted to generate an optimal debris removal plan to guide the mission process. According to the problem characteristics, a two-layer time-dependent traveling salesman problem(TSP) mathematical model is established, involving the debris removal sequence planning and the transfer trajectory planning. Subsequently, two main novel methods based on machine learning are proposed for the ADR mission planning problem, including a deep neural networks(DNN)-based estimation method for approximating the optimal velocity increments of perturbed multiple-impulse rendezvous and an reinforcement learning(RL)-based method for optimizing the sequence of debris removal and rendezvous time. Experimental results of different simulation scenarios have verified the effectiveness and superiority of the proposed method, indicating the good performance for solving the active debris removal mission planning problem. [ABSTRACT FROM AUTHOR]

Harmonic generation in power system networks presents significant issues that arise in power utilities. This paper describes a machine learning technique that was used to conduct a research study on the harmonic analysis of railway power stations. The research was an investigation of a time series whose values represented the total harmonic distortion (THD) for the electric current. This study was based on information collected at a railway power station. In an electrified substation, measurements of currents and voltages were made during a certain interval of time. From electric current values, the THD was calculated using a fast Fourier transform analysis (FFT) and the results were used to train an adaptive ANN—GMDH (artificial neural network–group method of data handling) algorithm. Following the training, a prediction model was created, the performance of which was investigated in this study. The model was based on the ANN—GMDH method and was developed for the prediction of the THD. The performance of this model was studied based on its parameters. The model's performance was evaluated using the regression coefficient (R), root-mean-square error (RMSE), and mean absolute error (MAE). The model's performance was very good, with an RMSE (root-mean-square error) value of less than 0.01 and a regression coefficient value higher than 0.99. Another conclusion from our research was that the model also performed very well in terms of the training time (calculation speed). [ABSTRACT FROM AUTHOR]

The clustering ensemble method has attracted much attention because it can improve the stability and robustness of single clustering methods. Among them, similarity-matrix-based methods or graph-based methods have had a wide range of applications in recent years. Most similarity-matrix-based methods calculate fully connected pairwise similarities by treating a base cluster as a whole and ignoring the importance of the relevance ranking of samples within the same base cluster. Since unreliable similarity estimates degrade clustering performance, constructing accurate similarity matrices is of great importance in applications. Higher-order graph diffusion based on reliable similarity matrices can further uncover potential connections between data. In this paper, we propose a more substantial graph-learning-based ensemble algorithm for local-sample-weighted clustering, which implicitly optimizes the adaptive weights of different neighborhoods based on the ranking importance of different neighbors. By further diffusion on the consensus matrix, we obtained an optimal consistency matrix with more substantial discriminative power, revealing the potential similarity relationship between samples. The experimental results showed that, compared with the second-best DREC algorithm, the accuracy of the proposed algorithm improved by 17.7%, and that of the normalized mutual information (NMI) algorithm improved by 15.88%. All empirical results showed that our clustering model consistently outperformed the related clustering methods. [ABSTRACT FROM AUTHOR]

The stochastic gradient descendant algorithm is one of the most popular neural network training algorithms. Many authors have contributed to modifying or adapting its shape and parametrizations in order to improve its performance. In this paper, the authors propose two modifications on this algorithm that can result in a better performance without increasing significantly the computational and time resources needed. The first one is a dynamic learning ratio depending on the network layer where it is applied, and the second one is a dynamic drop-out that decreases through the epochs of training. These techniques have been tested against different benchmark function to see their effect on the learning process. The obtained results show that the application of these techniques improves the performance of the learning of the neural network, especially when they are used together. [ABSTRACT FROM AUTHOR]

In recent years, finding the optimal solution for image segmentation has become more important in many applications. The whale optimization algorithm (WOA) is a metaheuristic optimization technique that has the advantage of achieving the global optimal solution while also being simple to implement and solving many real-time problems. If the complexity of the problem increases, the WOA may stick to local optima rather than global optima. This could be an issue in obtaining a better optimal solution. For this reason, this paper recommends a hybrid algorithm that is based on a mixture of the WOA and gray wolf optimization (GWO) for segmenting the brain sub regions, such as the gray matter (GM), white matter (WM), ventricle, corpus callosum (CC), and hippocampus (HC). This hybrid mixture consists of two steps, i.e., the WOA and GWO. The proposed method helps in diagnosing Alzheimer's disease (AD) by segmenting the brain sub regions (SRs) by using a hybrid of the WOA and GWO (H-WOA-GWO, which is represented as HWGO). The segmented region was validated with different measures, and it shows better accuracy results of 92%. Following segmentation, the deep learning classifier was utilized to categorize normal and AD images. The combination of WOA and GWO yields an accuracy of 90%. As a result, it was discovered that the suggested method is a highly successful technique for identifying the ideal solution, and it is paired with a deep learning algorithm for classification. [ABSTRACT FROM AUTHOR]

Bitcoin, the largest cryptocurrency, is extremely volatile and hence needs a better model for its pricing. In the literature, many researchers have studied the effect of data normalization on regression analysis for stock price prediction. How has data normalization affected Bitcoin price prediction? To answer this question, this study analyzed the prediction accuracy of a Legendre polynomial-based neural network optimized by the mutated climb monkey algorithm using nine existing data normalization techniques. A new dual normalization technique was proposed to improve the efficiency of this model. The 10 normalization techniques were evaluated using 15 error metrics using a multi-criteria decision-making (MCDM) approach called technique for order performance by similarity to ideal solution (TOPSIS). The effect of the top three normalization techniques along with the min–max normalization was further studied for Chebyshev, Laguerre, and trigonometric polynomial-based neural networks in three different datasets. The prediction accuracy of the 16 models (each of the four polynomial-based neural networks with four different normalization techniques) was calculated using 15 error metrics. A 16 × 15 TOPSIS analysis was conducted to rank the models. The convergence plot and the ranking of the models indicated that data normalization plays a significant role in the prediction capability of a Bitcoin price predictor. This paper can significantly contribute to the research with a new normalization technique for utilization in varied fields of research. It can also contribute to international finance as a decision-making tool for different investors as well as stakeholders for Bitcoin pricing. [ABSTRACT FROM AUTHOR]

The crowdsourcing system is an internet-based distributed problem-solving and production organization model, which has been applied in human–computer interaction, databases, natural language processing, machine learning and other fields. It guides the public to complete some tasks through specific strategies and methods. However, rational and selfish workers in crowdsourcing systems will submit solutions of different qualities in order to maximize their own benefits. Therefore, how to choose optimal strategies for selfish workers to maximize their benefits is important and crucial in such a scenario. In this paper, we propose a decision optimization method with incomplete information in a crowdsourcing system based on zero-determinant (ZD) strategies to help workers make optimal decisions. We first formulate the crowdsourcing problem, where workers have "winner-takes-all" rules as an iterated game with incomplete information. Subsequently, we analyze the optimal decision of workers in crowdsourcing systems in terms of ZD strategies, for which we find conditions to reach the maximum payoff of a focused worker. In addition, the analysis helps understand what solutions selfish workers will submit under the condition of having incomplete information. Finally, numerical simulations illustrate the performances of different strategies and the effects of the parameters on the payoffs of the focused worker. [ABSTRACT FROM AUTHOR]

It is well known that the one-dimensional cutting stock problem (1DCSP) is a combinatorial optimization problem with nondeterministic polynomial (NP-hard) characteristics. Heuristic and genetic algorithms are the two main algorithms used to solve the cutting stock problem (CSP), which has problems of small scale and low-efficiency solutions. To better improve the stability and versatility of the solution, a mathematical model is established, with the optimization objective of the minimum raw material consumption and the maximum remaining material length. Meanwhile, a novel algorithm based on deep reinforcement learning (DRL) is proposed in this paper. The algorithm consists of two modules, each designed for different functions. Firstly, the pointer network with encoder and decoder structure is used as the policy network to utilize the underlying mode shared by the 1DCSP. Secondly, the model-free reinforcement learning algorithm is used to train network parameters and optimize the cutting sequence. The experimental data show that the one-dimensional cutting stock algorithm model based on deep reinforcement learning (DRL-CSP) can obtain the approximate satisfactory solution on 82 instances of 3 data sets in a very short time, and shows good generalization performance and practical application potential. [ABSTRACT FROM AUTHOR]

In this paper, we study user grouping, precoding design, and power allocation for multiple-input multiple-output (MIMO) nonorthogonal multiple access (NOMA) systems. An optimization problem is formulated to the maximize the sum rate under a transmit power constraint at a base station and rate constraints on users, which are nonconvex and combinatorial and thus very challenging to solve. To tackle this problem, we carry out the optimization in two steps. In the first step, exploiting the machine learning technique, we develop an efficient iterative algorithm for user grouping and precoding design. In the second step, we develop a power-allocation scheme in closed form by recasting the original problem into a useful and tractable convex form. The numerical results demonstrate that the proposed joint scheme, including user grouping, precoding design, and power allocation, considerably outperforms the existing schemes in terms of sum rate maximization, which increases the sum-rate up to 8–18%. In addition, the results show the larger the number of antennas or users, the bigger the performance gap, at the cost of less computational complexity. [ABSTRACT FROM AUTHOR]

The design of analog circuits is a complex and repetitive process aimed at finding the best design variant. It is characterized by uncertainty and multivariate approaches. The designer has to make different choices to satisfy a predefined specification with required parameters. This paper proposes a method for facilitating the design of analog amplifiers based on m-polar fuzzy graphs theory and deep learning. M-polar fuzzy graphs are used because of their flexibility and the possibility to model different real-life multi-attribute problems. Deep learning is applied to solve a regression task and to predict the membership functions of the m-polar fuzzy graph vertices (the solutions), taking on the role of domain experts. The performance of the learner is high since the obtained errors are very small: Root Mean Squared Error is from 0.0032 to 0.0187, Absolute Error is from 0.022 to 0.098 and Relative Error is between 0.27% and 1.57%. The proposed method is verified through the design of three amplifiers: summing amplifier, subtracting amplifier, and summing/subtracting amplifier. The method can be used for improving the design process of electronic circuits with the possibility of automating some tasks. [ABSTRACT FROM AUTHOR]

Rockburst is a common and huge hazard in underground engineering, and the scientific prediction of rockburst disasters can reduce the risks caused by rockburst. At present, developing an accurate and reliable rockburst risk prediction model remains a great challenge due to the difficulty of integrating fusion algorithms to complement each other's strengths. In order to effectively predict rockburst risk, firstly, 410 sets of valid rockburst data were collected as the original data set in this paper, which was used to process these rockburst cases by the SMOTE oversampling method. Then, four integrated algorithms and eight basic algorithms were selected, which were optimized by hyperparameters and five-fold cross-validation and combined with the random search grid method, thus improving the classification performance of these algorithms. Third, the stacking integration algorithm, which was combined with the principles of various machine learning algorithms and the characteristics of the rockburst cases, integrated the optimization of rockburst algorithms with reference to four combinatorial strategies. Further, we adopted the voting integration algorithm, chose multiple combination schemes, and referred to the weighted fusion of accuracy, F1 score, recall, precision, and cv-mean as the weight values, and the optimal model for rockburst risk prediction was obtained. Finally, using the 35 generated stacking integration algorithms and 18 voting integration algorithms, the optimal model in the fusion strategy was selected and the traditional integration algorithm model was analyzed on the basis of different sample combinations of the models. The results showed that the prediction performance of stacking and voting integration algorithms was mostly better than the ordinary machine-learning performance, and the selection of appropriate fusion strategies could effectively improve the performance of rockburst prediction for ensemble learning algorithms. [ABSTRACT FROM AUTHOR]

This research is devoted to the determination of hidden dependencies between the flow of particles that come from the Sun and precipitation-induced floods in the United Kingdom (UK). The analysis covers 20 flood events during the period from October 2001 to December 2019. The parameters of solar activity were used as model input data, while precipitations data in the period 10 days before and during each flood event were used as model output. The time lag of 0–9 days was taken into account in the research. Correlation analysis was conducted to determine the degree of randomness for the time series of input and output parameters. For establishing a potential causative link, machine learning classification predictive modeling was applied. Two approaches, the decision tree, and the random forest were used. We analyzed the accuracy of classification models forecast from 0 to 9 days in advance. It was found that the most important factors for flood forecasting are proton density with a time lag of 9, differential proton flux in the range of 310–580 keV, and ion temperature. Research in this paper has shown that the decision tree model is more accurate and adequate in predicting the appearance of precipitation-induced floods up to 9 days ahead with an accuracy of 91%. The results of this study confirmed that by increasing technical capabilities, using improved machine learning techniques and large data sets, it is possible to improve the understanding of the physical link between the solar wind and tropospheric weather and help improve severe weather forecasting. [ABSTRACT FROM AUTHOR]

This paper deals with the problem of diagnosing oncological diseases based on blood protein markers. The goal of the study is to develop a novel approach in decision-making on diagnosing oncological diseases based on blood protein markers by generating datasets that include various combinations of features: both known features corresponding to blood protein markers and new features generated with the help of mathematical tools, particularly with the involvement of the non-linear dimensionality reduction algorithm UMAP, formulas for various entropies and fractal dimensions. These datasets were used to develop a group of multiclass kNN and SVM classifiers using oversampling algorithms to solve the problem of class imbalance in the dataset, which is typical for medical diagnostics problems. The results of the experimental studies confirmed the feasibility of using the UMAP algorithm and approximation entropy, as well as Katz and Higuchi fractal dimensions to generate new features based on blood protein markers. Various combinations of these features can be used to expand the set of features from the original dataset in order to improve the quality of the received classification solutions for diagnosing oncological diseases. The best kNN and SVM classifiers were developed based on the original dataset augmented respectively with a feature based on the approximation entropy and features based on the UMAP algorithm and the approximation entropy. At the same time, the average values of the metric MacroF 1 - score used to assess the quality of classifiers during cross-validation increased by 16.138% and 7.910%, respectively, compared to the average values of this metric in the case when the original dataset was used in the development of classifiers of the same name. [ABSTRACT FROM AUTHOR]

The Analytic Hierarchy Process (AHP) multicriteria method can be cognitively demanding for large-scale decision problems due to the requirement for the decision maker to make pairwise evaluations of all alternatives. To address this issue, this paper presents an interactive method that uses online learning to provide scalability for AHP. The proposed method involves a machine learning algorithm that learns the decision maker's preferences through evaluations of small subsets of solutions, and guides the search for the optimal solution. The methodology was tested on four optimization problems with different surfaces to validate the results. We conducted a one factor at a time experimentation of each hyperparameter implemented, such as the number of alternatives to query the decision maker, the learner method, and the strategies for solution selection and recommendation. The results demonstrate that the model is able to learn the utility function that characterizes the decision maker in approximately 15 iterations with only a few comparisons, resulting in significant time and cognitive effort savings. The initial subset of solutions can be chosen randomly or from a cluster. The subsequent ones are recommended during the iterative process, with the best selection strategy depending on the problem type. Recommendation based solely on the smallest Euclidean or Cosine distances reveals better results on linear problems. The proposed methodology can also easily incorporate new parameters and multicriteria methods based on pairwise comparisons. [ABSTRACT FROM AUTHOR]

In this paper, a stochastic configuration based fuzzy inference system with interpretable fuzzy rules (SCFS-IFRs) is proposed to improve the interpretability and performance of the fuzzy inference system and determine autonomously an appropriate model structure. The proposed SCFS-IFR first accomplishes a fuzzy system through interpretable linguistic fuzzy rules (ILFRs), which endows the system with clear semantic interpretability. Meanwhile, using an incremental learning method based on stochastic configuration, the appropriate architecture of the system is determined by incremental generation of ILFRs under a supervision mechanism. In addition, the particle swarm optimization (PSO) algorithm, an intelligence search technique, is used in the incremental learning process of ILFRs to obtain better random parameters and improve approximation accuracy. The performance of SCFS-IFRs is verified by regression and classification benchmark datasets. Regression experiments show that the proposed SCFS-IFRs perform best on 10 of the 20 data sets, statistically significantly outperforming the other eight state-of-the-art algorithms. Classification experiments show that, compared with the other six fuzzy classifiers, SCFS-IFRs achieve higher classification accuracy and better interpretation with fewer rules. [ABSTRACT FROM AUTHOR]

With the current shift in the mass media landscape from journalistic rigor to social media, personalized social media is becoming the new norm. Although the digitalization progress of the media brings many advantages, it also increases the risk of spreading disinformation, misinformation, and malformation through the use of fake news. The emergence of this harmful phenomenon has managed to polarize society and manipulate public opinion on particular topics, e.g., elections, vaccinations, etc. Such information propagated on social media can distort public perceptions and generate social unrest while lacking the rigor of traditional journalism. Natural Language Processing and Machine Learning techniques are essential for developing efficient tools that can detect fake news. Models that use the context of textual data are essential for resolving the fake news detection problem, as they manage to encode linguistic features within the vector representation of words. In this paper, we propose a new approach that uses document embeddings to build multiple models that accurately label news articles as reliable or fake. We also present a benchmark on different architectures that detect fake news using binary or multi-labeled classification. We evaluated the models on five large news corpora using accuracy, precision, and recall. We obtained better results than more complex state-of-the-art Deep Neural Network models. We observe that the most important factor for obtaining high accuracy is the document encoding, not the classification model's complexity. [ABSTRACT FROM AUTHOR]

The opportunity for large amounts of open-for-public and available data is one of the main drivers of the development of an information society at the beginning of the 21st century. In this sense, acquiring knowledge from these data using different methods of machine learning is a prerequisite for solving complex problems in many spheres of human activity, starting from medicine to education and the economy, including traffic as today's important economic branch. Having this in mind, this paper deals with the prediction of the risk of traffic incidents using both historical and real-time data for different atmospheric factors. The main goal is to construct an ensemble model based on the use of several machine learning algorithms which has better characteristics of prediction than any of those installed when individually applied. In global, a case-proposed model could be a multi-agent system, but in a considered case study, a two-agent system is used so that one agent solves the prediction task by learning from the historical data, and the other agent uses the real time data. The authors evaluated the obtained model based on a case study and data for the city of Niš from the Republic of Serbia and also described its implementation as a practical web citizen application. [ABSTRACT FROM AUTHOR]

Traveling Salesman Problems (TSPs) have been a long-lasting interesting challenge to researchers in different areas. The difficulty of such problems scales up further when multiple objectives are considered concurrently. Plenty of work in evolutionary algorithms has been introduced to solve multi-objective TSPs with promising results, and the work in deep learning and reinforcement learning has been surging. This paper introduces a multi-objective deep graph pointer network-based reinforcement learning (MODGRL) algorithm for multi-objective TSPs. The MODGRL improves an earlier multi-objective deep reinforcement learning algorithm, called DRL-MOA, by utilizing a graph pointer network to learn the graphical structures of TSPs. Such improvements allow MODGRL to be trained on a small-scale TSP, but can find optimal solutions for large scale TSPs. NSGA-II, MOEA/D and SPEA2 are selected to compare with MODGRL and DRL-MOA. Hypervolume, spread and coverage over Pareto front (CPF) quality indicators were selected to assess the algorithms' performance. In terms of the hypervolume indicator that represents the convergence and diversity of Pareto-frontiers, MODGRL outperformed all the competitors on the three well-known benchmark problems. Such findings proved that MODGRL, with the improved graph pointer network, indeed performed better, measured by the hypervolume indicator, than DRL-MOA and the three other evolutionary algorithms. MODGRL and DRL-MOA were comparable in the leading group, measured by the spread indicator. Although MODGRL performed better than DRL-MOA, both of them were just average regarding the evenness and diversity measured by the CPF indicator. Such findings remind that different performance indicators measure Pareto-frontiers from different perspectives. Choosing a well-accepted and suitable performance indicator to one's experimental design is very critical, and may affect the conclusions. Three evolutionary algorithms were also experimented on with extra iterations, to validate whether extra iterations affected the performance. The results show that NSGA-II and SPEA2 were greatly improved measured by the Spread and CPF indicators. Such findings raise fairness concerns on algorithm comparisons using different fixed stopping criteria for different algorithms, which appeared in the DRL-MOA work and many others. Through these lessons, we concluded that MODGRL indeed performed better than DRL-MOA in terms of hypervolumne, and we also urge researchers on fair experimental designs and comparisons, in order to derive scientifically sound conclusions. [ABSTRACT FROM AUTHOR]

A robot is essential in many industrial and manufacturing facilities due to its efficiency, accuracy, and durability. However, continuous use of the robotic system can result in various component failures. The servo motor is one of the critical components, and its bearing is one of the vulnerable parts, hence failure analysis is required. Some previous prognostics and health management (PHM) methods are very limited in considering the realistic operating conditions of industrial robots based on various operating speeds, loading conditions, and motions, because they consider constant speed data with unloading conditions. This paper implements a PHM for the servo motor of a robotic arm based on variable operating conditions. Principal component analysis-based dimensionality reduction and correlation analysis-based feature selection are compared. Two machine learning algorithms have been used to detect fault features under various operating conditions. This method is proposed as a robust fault-detection model for industrial robots under various operating conditions. Features from different domains not only improved the generalization of the model's performance but also improved the computational efficiency of massive data by reducing the total number of features. The results showed more than 90% accuracy under various operating conditions. As a result, the proposed method shows the possibility of robust failure diagnosis under various operating conditions similar to the actual industrial environment. [ABSTRACT FROM AUTHOR]

Traditional machine learning sequence models, such as RNN and LSTM, can solve sequential data problems with the use of internal memory states. However, the neuron units and weights are shared at each time step to reduce computational costs, limiting their ability to learn time-varying relationships between model inputs and outputs. In this context, this paper proposes two methods to characterize the dynamic relationships in real-world sequential data, namely, the internal time-varying sequence model (ITV model) and the external time-varying sequence model (ETV model). Our methods were designed with an automated basis expansion module to adapt internal or external parameters at each time step without requiring high computational complexity. Extensive experiments performed on synthetic and real-world data demonstrated superior prediction and classification results to conventional sequence models. Our proposed ETV model is particularly effective at handling long sequence data. [ABSTRACT FROM AUTHOR]

In this paper, a deep learning optimization algorithm is proposed, which is based on the Grünwald–Letnikov (G-L) fractional order definition. An optimizer fractional calculus gradient descent based on the G-L fractional order definition (FCGD_G-L) is designed. Using the short-memory effect of the G-L fractional order definition, the derivation only needs 10 time steps. At the same time, via the transforming formula of the G-L fractional order definition, the Gamma function is eliminated. Thereby, it can achieve the unification of the fractional order and integer order in FCGD_G-L. To prevent the parameters falling into local optimum, a small disturbance is added in the unfolding process. According to the stochastic gradient descent (SGD) and Adam, two optimizers' fractional calculus stochastic gradient descent based on the G-L definition (FCSGD_G-L), and the fractional calculus Adam based on the G-L definition (FCAdam_G-L), are obtained. These optimizers are validated on two time series prediction tasks. With the analysis of train loss, related experiments show that FCGD_G-L has the faster convergence speed and better convergence accuracy than the conventional integer order optimizer. Because of the fractional order property, the optimizer exhibits stronger robustness and generalization ability. Through the test sets, using the saved optimal model to evaluate, FCGD_G-L also shows a better evaluation effect than the conventional integer order optimizer. [ABSTRACT FROM AUTHOR]

Packing problems, also known as nesting problems or bin packing problems, are classic and popular NP-hard problems with high computational complexity. Inspired by classic reinforcement learning (RL), we established a mathematical model for two-dimensional (2D) irregular-piece packing combined with characteristics of 2D irregular pieces. An RL algorithm based on Monte Carlo learning (MC), Q-learning, and Sarsa-learning is proposed in this paper to solve a 2D irregular-piece packing problem. Additionally, mechanisms of reward–return and strategy-update based on piece packing were designed. Finally, the standard test case of irregular pieces was used for experimental testing to analyze the optimization effect of the algorithm. The experimental results show that the proposed algorithm can successfully realize packing of 2D irregular pieces. A similar or better optimization effect can be obtained compared to some classical heuristic algorithms. The proposed algorithm is an early attempt to use machine learning to solve 2D irregular packing problems. On the one hand, our hybrid RL algorithm can provide a basis for subsequent deep reinforcement learning (DRL) to solve packing problems, which has far-reaching theoretical significance. On the other hand, it has practical significance for improving the utilization rate of raw materials and broadening the application field of machine learning. [ABSTRACT FROM AUTHOR]

The increased usage of cyber-physical systems (CPS) has gained the focus of cybercriminals, particularly with the involvement of the internet, provoking an increased attack surface. The increased usage of these systems generates heavy data flows, which must be analyzed to ensure security. In particular, machine learning (ML) and deep learning (DL) algorithms have shown feasibility and promising results to fulfill the security requirement through the adoption of intelligence. However, the performance of these models strongly depends on the model structure, hyper-parameters, dataset, and application. So, the developers only possess control over defining the model structure and its hyper-parameters for diversified applications. Generally, not all models perform well in default hyper-parameter settings. Their specification is a challenging and complex task and requires significant expertise. This problem can be mitigated by utilizing hyper-parameter optimization (HPO) techniques, which intend to automatically find efficient learning model hyper-parameters in specific applications or datasets. This paper proposes an enhanced intelligent security mechanism for CPS by utilizing HPO. Specifically, exhaustive HPO techniques have been considered for performance evaluation and evaluation of computational requirements to analyze their capabilities to build an effective intelligent security model to cope with security infringements in CPS. Moreover, we analyze the capabilities of various HPO techniques, normalization, and feature selection. To ensure the HPO, we evaluated the effectiveness of a DL-based artificial neural network (ANN) on a standard CPS dataset under manual hyper-parameter settings and exhaustive HPO techniques, such as random search, directed grid search, and Bayesian optimization. We utilized the min-max algorithm for normalization and SelectKBest for feature selection. The HPO techniques performed better than the manual hyper-parameter settings. They achieved an accuracy, precision, recall, and F1 score of more than 98%. The results highlight the importance of HPO for performance enhancement and reduction of computational requirements, human efforts, and expertise. [ABSTRACT FROM AUTHOR]

This paper uses computational intelligence and machine learning methods to describe experimental modeling performed to approximate the static characteristics of one type of fluidic muscle from the manufacturer FESTO for three different muscle sizes. For the experiments, measured data from the manufacturer and data from a real system (i.e., test device) were used. The measurements, which took place on the experimental equipment, were carried out in two stages (i.e., when the muscle was pressed and when the muscle was relaxed). The resulting measured characteristics were obtained by averaging two values at a given moment. MATLAB® software was used for simulations, in which four models were created: MLP, SVM, ANFIS, and a custom model (i.e., polynomial model). Given that most articles mainly interpret their results graphically when approximating characteristics, in this article, the outputs of the models are also compared with the measured data based on the SSE, NRMSE, SBC, and AIC performance indicators, enabling a more relevant and comprehensive overview of the performance of the individual models. The outputs of the best models described in this article reach an accuracy of 89.90% to 98.74% (all from the MLP group), depending on the muscle size, compared to real measured outputs. [ABSTRACT FROM AUTHOR]

Digital twins are the most important parts of Cyber-Physical Systems (CPSs), and play a crucial role in the realization of the Metaverse. Therefore, two important factors: flexibility and adaptability, need to be focused on digital twinning systems. From a virtual perspective, constructing buildings, structures, and mechanisms in the Metaverse requires digital materials and components. Hence, accurate and reliable digital models can guarantee the success of implementation, particularly when it comes to completing physical twins in the real world. Accordingly, four Machine Learning (ML) methods to make digital twins of an advanced 42SiCr alloy considering all of its uncertainties and non-linearities have been employed in this paper. These ML methods accelerate the digitalization of the proposed alloy and allow users to employ them for a wide range of similar metals. Based on this technique, producers can borrow these virtual materials and build their structures in the Metaverse. This way, if the properties of the materials were satisfactory, they might buy them and start manufacturing their products. As a case study, we focus on digital twining of an 42SiCr steel with some influential factors in its mechanical properties, making the nature of the alloy complex. Processes, including heat treatment, may restore the material's deformability; however, Quenching and Partitioning (Q&P) not only eliminates the impact of cold forming but also provides advanced high-strength steel (AHSS) properties. In this research, the combined impacts of different Q&P treatments were investigated on the mechanical properties of 42SiCr steel alloy. The results have shown the acceptability and accuracy of the proposed ML methods in realizing the digital twins of this complex alloy. [ABSTRACT FROM AUTHOR]