Search

This paper investigates a online learning optimization problem in a distributed manner, where a set of agents aim to cooperatively minimize the sum of local time-varying cost functions while communications among agents depend on a sequence of time-varying directed graphs. For such a problem, we first propose a online distributed stochastic (sub)gradient push-sum algorithm by utilizing distributed optimization methods and push-sum protocols. Then we analyze the regret bounds for the proposed algorithm on the cases when the cost functions are convex and strongly convex, respectively. The bound on the expected regret for convex functions grows sub-linearly with order of O (T) , where T is the time horizon. When the cost functions are strongly convex with Lipschitz gradients, the regret bound has an improved rate with order of O (ln T). Numerical simulations on the localization in wireless sensor networks are used to show the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Deep learning is a branch of neural network which has been intensively developed in the last decade. Due to the high-accuracy classification ability, the deep learning algorithms have been widely used in many fields, such as speech recognition, image recognition, and natural speech processing. However, they also show some shortcomings especially on the selection of some parameters in the network, including hyper-parameters, which is still treated as a time consuming task. In this paper, a modified ABC (ABC-ISB) optimization algorithm is proposed to automatically train the parameters of Feed-Forward Artificial Neural Networks, which is a typical a neural network. In the proposed ABC algorithm, we utilize the information of neighbors with better performance to accelerate the convergence of employed and onlooker bees respectively. In addition, a new selection strategy and a gbest-guided strategy are introduced to enhance the global search capability and balance the exploration and exploitation of the algorithm separately. The experimental results show our ABC-ISB is generally leading and competitive. [ABSTRACT FROM AUTHOR]

Hyperparameter optimization is a challenging process that has the potential to improve machine learning algorithms. Since it creates a remarkable computational burden for machine learning tasks, there have been few works coping with tuning strategies of a specific algorithm. In this paper, an improved Stochastic Gradient Descent (SGD) based on Fisher Maximization is developed for tuning hyperparameters of an Echo State Network (ESN) which has a wide range of applications. The results of the method are then compared with those of traditional Gradient Descent and Grid Search. According to the obtained results; 1) The scale of the data sets greatly affects the reliability of hyperparameter optimization results; 2) Feature selection is critical in terms of mean error of training when hyperparameter optimization is applied on some methods such as ESN; 3) SGD falls in a good local minima if Fisher Maximization is performed to find a good starting point. [ABSTRACT FROM AUTHOR]

• A comprehensive review of LL fault diagnosis methods in PV systems. • An ensemble learning model based on probabilistic strategy is developed. • The main features of the fault are extracted from I-V curves using a Simulink based model of PV arrays. • A feature selection algorithm is used for each learning algorithm to reduce the dataset and increase the classification accuracy. • The proposed model is very reliable and able to detect and classify all the LL faults accurately. The fault diagnosis of photovoltaic (PV) arrays aims to increase the reliability and service life of PV systems. Line-Line (LL) faults may remain undetected under low mismatch level and high impedance due to low currents of faults, resulting in power losses and fire potential disaster. This paper proposes a novel and intelligent fault diagnosis method based on an ensemble learning model and Current-Voltage (I-V) characteristics to detect and classify LL faults at the DC side of PV systems. For this purpose, first, the key features are extracted via analyzing I-V characteristics under various LL fault events and normal operation. Second, a feature selection algorithm has been applied to select the best features for each learning algorithm in order to reduce the amount of data required for the learning process. Third, an ensemble learning model is developed that combines several learning algorithms based on the probabilistic strategy to achieve superior diagnostic performance. Here, we find an excellent agreement between simulation and experimental results that the proposed method can obtain higher accuracy in detecting and classifying the LL faults, even under low mismatch levels and high fault impedances. In addition, the comparison results demonstrate that the performance of the proposed method is better than individual machine learning algorithms, so that the proposed method precisely detects and classifies LL faults on PV systems under the different conditions with an average accuracy of 99% and 99.5%, respectively. [ABSTRACT FROM AUTHOR]

The free-flying space manipulators present challenges in controlling the motions of both the spacecraft bus and the manipulator, because of the highly-coupling system dynamics and the unknown space environment disturbances. Although the sliding mode controllers are robust to the unknown disturbances and system uncertainties, the chattering effect could affect the pointing accuracy and the lifetime of the actuators. This paper first introduces the dynamics of a CuBot, which is a 3-rigid-link manipulator based on the CubeSat platform. To maintain the robustness while decreasing the chattering effect, an innovative reinforcement learning based fuzzy adaptive sliding mode controller is proposed. To maintain the robustness while reducing the chattering effect, an innovative reinforcement learning based fuzzy adaptive sliding mode controller is proposed. The switching gain is updated to estimate the lumped upper bound of the system uncertainties and the unknown disturbances, and then a new fuzzy logic adaptive law is applied on the switching gain to decrease the chattering effects. On top of that, the fuzzy logic rules are tuned by an innovative modified reinforcement learning mechanism to achieve the better tracking performance. The uniformly ultimately bounded tracking errors are guaranteed by the proposed control scheme, and the effectiveness is validated by the simulation results. • A space manipulator is designed on the 8U CubeSat bus. • A modified reinforcement learning algorithm is designed to tune the fuzzy logic rules. • A fuzzy adaptive sliding mode controller is designed to reduce the chattering effect. • The new algorithm is compared to a conventional sliding mode controller. • The performance of the new algorithm is verified by simulation. [ABSTRACT FROM AUTHOR]

• BNGBS is proposed to pursue an efficient learning process for classification task. • Triple incremental abilities are developed to learn more nodes, samples, and classes. • Extensive experiments are conducted on eight datasets to validate the effectiveness. As an ensemble algorithm, network boosting enjoys a powerful classification ability but suffers from the tedious and time-consuming training process. To tackle the problem, in this paper, a broad network gradient boosting system (BNGBS) is developed by integrating gradient boosting machine with broad networks, in which the classification loss caused by a base broad network is learned and eliminated by followed networks in a cascade manner. The proposed system is constructed as an additive model and can be easily optimized by a greedy strategy instead of the tedious back-propagation algorithm, resulting in a more efficient learning process. Meanwhile, triple incremental learning capabilities including the increment of feature nodes, increment of input samples, and increment of target classes are designed. The proposed system can be efficiently updated and expanded based on the current status instead of being entirely retrained when the demands for more feature nodes, input samples, and target classes are proposed. The node-increment ability allows to add more feature nodes into the built system if the current structures are not effective for learning. The sample-increment ability is developed to allow the model to keep learning from the coming batch data. The class-increment ability is used to tackle the issue that the coming batch data may contain unseen categories. In comparison with existing popular machine learning methods, comprehensive results based on eight benchmark datasets illustrate the effectiveness of the proposed broad network gradient boosting system for the classification task. [ABSTRACT FROM AUTHOR]

We explored the problem about function approximation error and complex mission adaptability in multi-agent deep reinforcement learning. This paper proposes a new multi-agent deep reinforcement learning algorithm framework named multi-agent time delayed deep deterministic policy gradient. Our work reduces the overestimation error of neural network approximation and variance of estimation result using dual-centered critic, group target network smoothing and delayed policy updating. According to experiment results, it improves the ability to adapt complex missions eventually. Then, we discuss that there is an inevitable overestimation issue about existing multi-agent algorithms about approximating real action-value equations with neural network. We also explain the approximate error of equations in the multi-agent deep deterministic policy gradient algorithm mathematically and experimentally. Finally, the application of our algorithm in the mixed cooperative competition experimental environment further demonstrates the effectiveness and generalization of our algorithm, especially improving the group's ability of adapting complex missions and completing more difficult missions. [ABSTRACT FROM AUTHOR]

Facial landmark tracking often adopts static models to generically fit per frame of video. This is considered inappropriate since such models ignore the informative correlation between previous and current frames. Moreover, most of these methods fail to balance the speed and accuracy for video-based facial landmark tracking simultaneously. In this paper, we propose an efficient online framework for video-based face alignment, named Incrementally Cascaded Broad Learning framework (ICBL). ICBL aims to continuously enhance the prediction capability of tracking model for sequential data. It is capable of learning the spatial appearance on specific-person statistics from continuous facial frames and using such knowledge to incrementally tune a cascade of regressors in parallel. To achieve this goal, we approximate the facial shape space by sampling from a dynamic distribution which is continuously updated by person-specific statistics from the tracked facial frames. This dramatically facilitates cascade regression to incrementally update all cascade-regressors in parallel, thus allowing a fast update of the whole model. Furthermore, we successfully incorporate both the linear and non-linear mappings into our parallel cascade framework and introduce Broad Learning (BL) algorithm as a solution for them simultaneously. Experimental results on the most popular and large-scale benchmark for facial landmark tracking show highly competitive performance of proposed ICBL in comparisons with the state-of-the-arts. The code of our ICBL framework has been available from https://github.com/CaifengLiu/Facial-landmark-tracking-by-ICBL. [ABSTRACT FROM AUTHOR]

Hyperparameter tuning is critical for the performance of machine learning algorithms. However, a noticeable limitation is the high computational cost of algorithm evaluation for complex models or for large datasets, which makes the tuning process highly inefficient. In this paper, we propose a novel model-based method for efficient hyperparameter optimization. Firstly, we frame this optimization process as a reinforcement learning problem and then employ an agent to tune hyperparameters sequentially. In addition, a model that learns how to evaluate an algorithm is used to speed up the training. However, model inaccuracy is further exacerbated by long-term use, resulting in collapse performance. We propose a novel method for controlling the model use by measuring the impact of the model on the policy and limiting it to a proper range. Thus, the horizon of the model use can be dynamically adjusted. We apply the proposed method to tune the hyperparameters of the extreme gradient boosting and convolutional neural networks on 101 tasks. The experimental results verify that the proposed method achieves the highest accuracy on 86.1% of the tasks, compared with other state-of-the-art methods and the average ranking of runtime is significant lower than all methods by using the predictive model. [ABSTRACT FROM AUTHOR]

Sparse representation and dictionary learning have been successfully applied in hyperspectral image classification. Generally, it is more effective to learn the sub-dictionary for each class and utilize multiple scale strategy. However, the sub-dictionary may only consider the within-class information and ignore the discriminative information. For multiscale sparse representation, the shape of the regions would not adaptively change according to context structure. This paper proposes a discriminative sub-dictionary learning algorithm and an adaptive multiscale superpixel classification strategy under sparse representation framework for hyperspectral image classification. In dictionary learning stage, the global constraint term can ensure the learnt sub-dictionaries contain more discriminative information. In sparse representation stage, the adaptive multiscale superpixel strategy can exploit the spatial context according to the structural information. The combination of the two strategies can further explore the spatial and spectral information in hyperspectral image. Experiments on four hyperspectral image datasets prove that the proposed method has a significant improvement in both quantitative and qualitative analysis comparing with state-of-the-art algorithms. Moreover, the proposed algorithm can produce satisfactory results for the imbalanced data problem involved in hyperspectral image. [ABSTRACT FROM AUTHOR]

This paper presents efficient algorithms for learning low-coherence dictionaries. First, a new algorithm based on proximal methods is proposed to solve the dictionary learning (DL) problem regularized with the mutual coherence of dictionary. This is unlike the previous approaches that solve a regularized problem where an approximate incoherence promoting term, instead of the mutual coherence, is used to encourage low-coherency. Then, a new solver is proposed for constrained low-coherence DL problem, i.e., a DL problem with an explicit constraint on the mutual coherence of the dictionary. As opposed to current methods, which follow a suboptimal two-step approach, the new algorithm directly solves the associated DL problem. Using previous studies, convergence of the new schemes to critical points of the associated cost functions is also provided. Furthermore, it is shown that the proposed algorithms have lower iteration complexity than existing algorithms. Our simulation results on learning low-coherence dictionaries for natural image patches as well as image classification based on discriminative over-complete dictionary learning demonstrate the superiority of the proposed algorithms compared with the state-of-the-art method. [ABSTRACT FROM AUTHOR]

Machine learning is increasingly used in mental health research and has the potential to advance our understanding of how to characterize, predict, and treat mental disorders and associated adverse health outcomes (e.g., suicidal behavior). Machine learning offers new tools to overcome challenges for which traditional statistical methods are not well-suited. This paper provides an overview of machine learning with a specific focus on supervised learning (i.e., methods that are designed to predict or classify an outcome of interest). Several common supervised learning methods are described, along with applied examples from the published literature. We also provide an overview of supervised learning model building, validation, and performance evaluation. Finally, challenges in creating robust and generalizable machine learning algorithms are discussed. [ABSTRACT FROM AUTHOR]

This paper proposes a novel domain adaptation algorithm to handle the challenges posed by the satellite and aerial imagery, and demonstrates its effectiveness on the built-up region segmentation problem. Built-up area estimation is an important component in understanding the human impact on the environment, effect of public policy and in general urban population analysis. The diverse nature of aerial and satellite imagery (capturing different geographical locations, terrains and weather conditions) and lack of labeled data covering this diversity makes machine learning algorithms difficult to generalize for such tasks, especially across multiple domains. Re-training for new domain is both computationally and labor expansive mainly due to the cost of collecting pixel level labels required for the segmentation task. Domain adaptation algorithms have been proposed to enable algorithms trained on images of one domain (source) to work on images from other dataset (target). Unsupervised domain adaptation is a popular choice since it allows the trained model to adapt without requiring any ground-truth information of the target domain. On the other hand, due to the lack of strong spatial context and structure, in comparison to the ground imagery, application of existing unsupervised domain adaptation methods results in the sub-optimal adaptation. We thoroughly study limitations of existing domain adaptation methods and propose a weakly-supervised adaptation strategy where we assume image level labels are available for the target domain. More specifically, we design a built-up area segmentation network (as encoder-decoder), with image classification head added to guide the adaptation. The devised system is able to address the problem of visual differences in multiple satellite and aerial imagery datasets, ranging from high resolution (HR) to very high resolution (VHR), by investigating the latent space as well as the structured output space. A realistic and challenging HR dataset is created by hand-tagging the 73.4 sq-km of Rwanda, capturing a variety of build-up structures over different terrain. The developed dataset is spatially rich compared to existing datasets and covers diverse built-up scenarios including built-up areas in forests and deserts, mud houses, tin and colored rooftops. Extensive experiments are performed by adapting from the single-source domain datasets, such as Massachusetts Buildings Dataset, to segment out the target domain. We achieve high gains ranging 11.6–52% in IoU over the existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper presents a machine learning based model for control of local bioaerosol concentration via a forced corner flow with optimal energy efficiency in an indoor environment. A recirculation zone determined by the inlet flow rate traps particles partially with one or more vortices around the corner. The profile of the recirculation zone is then determined mathematically by the minimum net mass flux principle with a grid search technique. Subsequently, the variation of the recirculation zone profile is then learned through a neural network (NN), in which data is collected from the simulation by the Eulerian–Lagrangian scheme. Moreover, a model predictive control (MPC) algorithm is implemented to achieve an optimal profile of the recirculation zone with optimal energy consumption, based on the linearized NN model. Finally, the proposed NN-MPC is implemented for simulation of removing the local bioaerosol from an indoor corner through a flow-rate-controllable airflow from ventilation outlet located on the ceiling. • A recirculation zone trapping particles is defined by minimum net mass flux law. • Relationship between recirculation zone and inlet flow rate is trained by NN. • Energy consumption of input is optimized to fit long-term continuous process. • The proposed NN-MPC reduces particle concentration with energy efficiency. [ABSTRACT FROM AUTHOR]

Feature selection (FS) is an important pre-processing step in data mining and pattern recognition. It can effectively compress the dimensionality of the feature space to reduce computation time and improve classification performance. The meta-heuristic algorithm-based feature selection method by finding the optimal set of features in the solution space has been widely used. However, this method is prone to trap into local optimality in a sufficiently large solution space. In this paper, we first propose a binary dandelion algorithm (BDA) to improve classification accuracy. In addition, to improve the performance of the algorithm, a binary dandelion algorithm using an improved seeding strategy and chaotic populations (SBDA) is proposed in this paper. Firstly, the strategy of optimizing the seeding radius by using the vibrational function and the historical optimal population increases the complexity of the search process and improves the search performance of the algorithm in the solution space. Secondly, when generating seeds, chaotic populations are generated using chaotic operators, which improves the ability of the algorithm to jump out of the local optimum and improves the stability of the algorithm. In this paper, 15 well-established datasets collected from the UCI machine learning database were adopted to compare four variants of BDA using only chaotic population improvement and in the next experiments, both mechanisms are verified to be effective in improving the performance of the algorithm. In addition, this paper compares the proposed BDA algorithm and SBDA algorithm with eight other classical algorithms. The experimental results show that SBDA can obtain fewer features with higher classification accuracy in most cases. • Proposed a BDA algorithm for solving binary problems. • Proposed a SBDA algorithm by incorporating chaos mapping. • Presented an improved seeding strategy in relation to the historical optimal population. • Studied and applied the feature selection problem with extreme learning machine. [ABSTRACT FROM AUTHOR]

This paper introduces a new hybrid bio-inspired solver which combines elements from the recently proposed Coral Reefs Optimization (CRO) algorithm with operators from the Harmony Search (HS) approach, which gives rise to the coined CRO-HS optimization technique. Specifically, this novel bio-inspired optimizer is utilized in the context of short-term wind speed prediction as a means to obtain the best set of meteorological variables to be input to a neural Extreme Learning Machine (ELM) network. The paper elaborates on the main characteristics of the proposed scheme and discusses its performance when predicting the wind speed based on the measures of two meteorological towers located in USA and Spain. The good results obtained in these experiments when compared to naïve versions of the CRO and HS algorithms are promising and pave the way towards the utilization of the derived hybrid solver in other optimization problems arising from diverse disciplines. [ABSTRACT FROM AUTHOR]

DVR is one of the custom power devices for compensating power quality indices. A self-tuning controller with a bi-objective structure is presented for controlling the DVR compensator in order to improve the THD and voltage sag indices of a sensitive load in the network. In this paper, the emotional controller which is based on emotional learning of human brain is proposed for controlling the DVR compensator. This controller has such a structure that makes it capable of considering a second objective in the control process of the system. So far, this capability of the emotional controller has not been used in any researches. The results of the paper demonstrate that compensating and controlling the voltage THD signal in the control process has caused more improvement in the voltage sag of the sensitive load. It was reported that the performance of the emotional controller depends on the selection of the values of its coefficients. Therefore, in order to better improve the proposed controller, these coefficients are tuned by an optimization algorithm. Teaching–learning-based optimization algorithm is considered as optimization algorithm to regulate these coefficients. According to simulation results, it works significantly better than classic PI controller and some intelligent controllers that have introduced in other researches already. [ABSTRACT FROM AUTHOR]

Because of the sparse reward and the sequence of the complex multi-step task, there is a big challenge in reinforcement learning, that is, the agent needs to implement several consecutive sequential steps to complete the whole task without the intermediate reward. Reward Shaping and Curriculum Learning algorithms are always used to solve this challenge, but Reward Shaping is prone to sub-optimal policies and Curriculum Learning easily suffers from the catastrophic forgetting problem. In this paper, we propose a novel algorithm called Parallel Multi-Environment Shaping (PMES), where several sub-environments are built based on human knowledge to make the agent aware of the importance of intermediate steps, each of which corresponds to a key intermediate step. Specifically, the learning agent is trained under these parallel multiple environments including the original environment and several sub-environments by synchronous advantage actor-critic algorithm. And the PMES algorithm has the mechanism of adaptive reward shaping to adjust the reward function. In this way, PMES algorithm effectively incorporates human experience by multiple different environments rather than only shaping the reward function, which combines the benefits of Reward Shaping and Curriculum Learning algorithms while avoiding their drawbacks. Extensive experiments on the mini-game 'Build Marines' of StarCraft II environment show that our proposed algorithm is more effective than Reward Shaping, Curriculum Learning, and PLAID algorithms, which is almost close to the level of human Grandmaster. And compared with the existing work, it takes less time and computing resources to reach a good result. [ABSTRACT FROM AUTHOR]

This paper presents an adaptive neuro-fuzzy PID controller based on twin delayed deep deterministic policy gradient (TD3) algorithm for nonlinear systems. In this approach, the observation of the environment is embedded with information of a multiple input single output (MISO) fuzzy inference system (FIS) and have a specially defined fuzzy PID controller in neural network (NN) formation acting as the actor in the TD3 algorithm, which achieves automatic tuning of gains of fuzzy PID controller. From the control perspective, the controller combines the merits of both FIS and PID controller and utilizes reinforcement learning algorithm for optimizing parameters. From the reinforcement learning point of view, embedding the prior knowledge into the fuzzy PID controller incorporated in the actor network helps reduce the learning difficulty in the training process. The proposed method was tested on the cart-pole system in simulation environment with comparison of a linear PID controller, which demonstrates the robustness and generalization of the proposed approach. [ABSTRACT FROM AUTHOR]

• A new graph-based data classification method is presented. • The method can be applied to both semi-supervised and unsupervised learning tasks. • Some of its properties allow it to perform well with small labeled training sets. • It is unconditionally stable, incorporates class size information and efficient. • Experiments are performed on classification of 3D sensory data such as point clouds. Data classification, where the goal is to divide data into predefined classes, is a fundamental problem in machine learning with many applications, including the classification of 3D sensory data. In this paper, we present a data classification method which can be applied to both semi-supervised and unsupervised learning tasks. The algorithm is derived by unifying complementary region-based and edge-based approaches; a gradient flow of the optimization energy is performed using modified auction dynamics. In addition to being unconditionally stable and efficient, the method is equipped with several properties allowing it to perform accurately even with small labeled training sets, often with considerably fewer labeled training elements compared to competing methods; this is an important advantage due to the scarcity of labeled training data. Some of the properties are: the embedding of data into a weighted similarity graph, the in-depth construction of the weights using, e.g., geometric information, the use of a combination of region-based and edge-based techniques, the incorporation of class size information and integration of random fluctuations. The effectiveness of the method is demonstrated by experiments on classification of 3D point clouds; the algorithm classifies a point cloud of more than a million points in 1–2 min. [ABSTRACT FROM AUTHOR]

• Spatiotemporal usage patterns of station-free shared bikes are explored. • A topic-based two-stage algorithm is proposed to discover city functional regions. • Region functions are labeled by static POI data and dynamic mobility patterns. Station-free shared bike (SFSB) is a new travel mode that shared bikes are allowed to park in any proper places. It implies that the users are more likely to park the SFSB as close as their destinations. This advantage makes the SFSB data to be an ideal source to understand the land-use distribution. Inspired by the idea in text mining, this paper proposes a topic-based two-stage SFSB data mining algorithm to understand the SFSB user's travel behavior and to discover city functional regions. A region, a function and human mobility patterns are treated as a document, a topic and words, respectively. Then, a region is represented by a distribution of functions, and a function is featured by a distribution of mobility patterns. The point-of-interest data is combined to annotate the clustered regions to discover the real functions. At last, the proposed method is tested using 14-day SFSB data in Beijing and the results are estimated by the Satellite Map data. The proposed methods and the results can be applied to infer the individual's travel purpose through functional regions and to improve land-use planning, etc. [ABSTRACT FROM AUTHOR]

This paper presents a new method for recognition of nine control chart patterns (CCPs) based on the intelligent use of shape and statistical features and optimized fuzzy system. The proposed technique contains three levels of separation. In each level of separation, an effective set of shape and statistical features are utilized as the input of classifier for recognizing a part of patterns. Due to the good performance of the adaptive neuro-fuzzy inference system (ANFIS) in pattern recognition problems, in the proposed method an ANFIS is used as a classifier at each level of separation which is trained by chaotic whale optimization algorithm (CWOA). Intelligent utilization of new extracted features, improving robustness of ANFIS and considering nine patterns in CCP recognition problem are the main contribution of the proposed method. The simulation results showed that the proposed method performs better than other similar methods and can recognize the type of pattern with 99.77% accuracy. • Intelligent application of features has high effect on classifier's performance. • Learning algorithm of ANFIS is an important issue that affects its accuracy. • The proposed method can recognize nine type of CCPs. • The results showed that the proposed method performs better than other methods. [ABSTRACT FROM AUTHOR]

In this paper, we propose an evolutionary learning approach for flexible identification of custom composite models for classification problems. To solve this problem in an efficient way, the problem-specific evolutionary operators are proposed and the effectiveness of different modifications of the common genetic programming algorithm is investigated. Also, several implementations of caching for the fitted models were compared from the performance point of view. To verify the proposed algorithm, both synthetic and real-world classification cases are examined. The implemented solution can identify the structure of the composite models from scratch, as well as be used as a part of automated machine learning solutions. [ABSTRACT FROM AUTHOR]

The article describes the automated formatting verification technique of docx document elements which were the basis for developed online service. Checking the document formatting correctness is an important task when writing various research papers, explanatory notes for course projects, and other scientific works. The article describes the approach for identifying design features of text document elements. The structure of the client-server interaction of the service, the operation of which was simulated, is also given. Such a machine learning algorithm as gradient boosting on decision trees CatBoost was chosen as the primary tool for multi-classification. Empirically refined parameters of the algorithm that increase its accuracy are described. Special attention is paid to the developed method of checking the results of classification of elements, the sequence of which has different regularities. This approach allows us to identify the formatting classes of docx files elements that were incorrectly identified by the classifier. Sometimes, it is possible to override the results of the classifier to increase the accuracy of checking the elements formatting in docx files. [ABSTRACT FROM AUTHOR]

In this paper, we propose a fuzzification method for clusters produced from a clustering process, based on Active Learning Method (ALM). ALM is a soft computing methodology which is based on a hypothesis claiming that human brain interprets information in pattern-like images. The proposed fuzzification method is applicable to all non-fuzzy clustering algorithms as a post process. The most outstanding advantage of this method is the ability to determine the membership degrees of each data to all clusters based on the density and shape of the clusters. It is worth mentioning that for existing fuzzy clustering algorithms such as FCM the membership degree is usually determined as a function of distance to the center of the clusters. In our proposed method, all data points of a cluster will play a role in order to determine the membership degrees. Consequently, the obtained membership degrees will depend on all of the data points of clusters, the amount of data points, and the density distribution of the clusters. Simulations prove the advantages of the proposed method. [ABSTRACT FROM AUTHOR]

Graph analytics are now considered the state-of-the-art in many applications of communities detection. The combination between the graph's definition in mathematics and the graphs in computer science as an abstract data structure is the key behind the success of graph-based approaches in machine learning. Based on graphs, several approaches have been developed such as shortest path first (SPF) algorithms, subgraphs extraction, social media analytics, transportation networks, bioinformatic algorithms, etc. While SPF algorithms are widely used in optimization problems, Spectral clustering (SC) algorithms have overcome the limits of the most state-of-art approaches in communities detection. The purpose of this paper is to introduce a graph-based approach of communities detection in the novel coronavirus Covid-19 countries' datasets. The motivation behind this work is to overcome the limitations of multiclass classification, as SC is an unsupervised clustering algorithm, there is no need to predefine the output clusters as a preprocessing step. Our proposed approach is based on a previous contribution on an automatic estimation of the k number of the output clusters. Based on dynamic statistical data for more than 200 countries, each cluster is supposed to group countries having similar behaviors of Covid-19 propagation. [ABSTRACT FROM AUTHOR]

The article deals with the subject of mini-models (MMs) based on clustering algorithms. The mini-model method is a local regression algorithm that operates on some part of the input space called the mini-model domain (MM domain). MM domain can be created as a multidimensional polytope in the input space. Another possible solution is to divide the input space with clustering algorithms. As a result of this process, each data cluster is treated as a separate mini-model domain. The main aim of the article is to create an exhaustive comparison of mini-model methods based on the most well-known clustering algorithms. The work introduces new versions of the mini-model method based on clustering algorithms such as DBSCAN, OPTICS, Mean Shift, spectral clustering and several hierarchical methods. The paper also compares the results with other versions of the MM-method and instance-based learning algorithms. [ABSTRACT FROM AUTHOR]

Intrusion detection has been the subject of numerous studies in industry and academia. Still, Cybersecurity analysts always want greater precision and global threat analysis to secure their systems in cyberspace. To improve the intrusion detection system, the visualization of the security events in the form of graphs and diagrams is significant to improve the accuracy of alerts. In this paper, we propose an approach of an IDS based on cloud computing, big data technique and, using a machine learning graph algorithm which can detect in real-time different attacks as early as possible. We use the MAWILab intrusion detection dataset. We choose Microsoft Azure as a unified cloud environment to load our dataset on Azure blob storage. We implement the k2 algorithm, which is a graphical machine learning algorithm to classify attacks. Our system showed a great performance due to the graphical machine learning algorithm and Apache Spark structured streaming engine. [ABSTRACT FROM AUTHOR]

In this paper an Online Monitoring System is presented, which enables a universal condition monitoring of electrical household appliances. The approach involves recording the power consumption of the respective device and evaluating the measured data using machine learning methods and tools. Thereby an artificial intelligence learns the normal condition of the examined device and can subsequently perform a monitoring. A k -Means algorithm is used for clustering and a One-class Support Vector Machine for classification. This makes it possible to detect anomalies in the operating modes and behavior of the device during subsequent operation. In addition to the methodology, the design and real application of a prototype is presented, in which a suitable sensor technology and a software with artificial intelligence are implemented. [ABSTRACT FROM AUTHOR]

This paper presents the design and a prototypical implementation of a tool for short-term energy demand forecasting in prosumer scenarios in local energy systems. The prototype combines explainable machine learning and visual analytics to facilitate the forecasting and analysis of energy demand and supply in a way usable for small utilities not used to complex analytical approaches. It applies a kNN (k-Nearest Neighbour) algorithm to forecast energy demand and supply and presents the results in an interactive visual dashboard that allows the user to analyze different forecast alternatives, to understand how they relate to the input parameters and identify consumption and production patterns. Initial assessment of the forecasting accuracy (MAPE 5.77%) on a limited dataset and the feedback from target users support the feasibility of the proposed approach. [ABSTRACT FROM AUTHOR]

Failure of machinery and equipment in factories, plants, etc. is a great loss to the operator. Various studies have been conducted for a long time to solve this problem, and many solutions have recently been developed to perform real-time health monitoring by attaching sensors to facilities. In addition, machine learning has allowed more accurate diagnosis of the condition to be carried out on these collected data. However, real-time fault detection is accompanied by a number of difficulties. Class imbalance issues are considered one of the most representative factors that makes it difficult to apply machine learning models in real-world locations. Class imbalance means that when it comes to classification problems, the amount of data in one class is greater than that in another class. In this case, the machine learning model is overfitted to a class with a lot of data, causing performance degradation. In real-world sites, the performance degradation of the model due to class imbalance is very serious because most data occurs in normal state, while fault data does not occur very well. Several prior studies have proposed ways to overcome this class imbalance, but each has several limitations. In this paper, an algorithm solution method and computing architecture solution method are introduced together as methodologies to overcome class imbalance, and a robust classification model is designed for class imbalance. [ABSTRACT FROM AUTHOR]

• The integer-valued vector representing a candidate solution encodes all dataset features. • The feature subset size is determined by using an additional element in the vector. • Feature relevance is represented through its relative position in the vector. • A permutation-matrix-based mutation operator can be used to create new solutions. • A simple repair-based recombination operator is used to ensure feasible solution creation. This paper describes a permutational-based Differential Evolution algorithm implemented in a wrapper scheme to find a feature subset to be applied in the construction of a near-optimal classifier. In this approach, the relevance of a feature chosen to build a better classifier is represented through its relative position in an integer-valued vector, and by using a permutational-based mutation operator, it is possible to create new feasible candidate solutions only. Furthermore, to provide a controlled diversity rate in the population, a straightforward repair-based recombination operator is utilized to evolve a population of candidate solutions. Unlike the other approaches in the existing literature using integer-valued vectors and requiring a predefined subset size, in this approach, this size is determined by an additional element included in the encoding scheme, allowing to find an adequate feature subset size to each specific dataset. Experimental results show that this approach is an effective way to create more accurate classifiers as they are compared with those obtained by other similar approaches. [ABSTRACT FROM AUTHOR]

As a popular traditional Chinese exercise, Tai Chi is of great benefit to physical and mental health. Professional and face-to-face guidance is a traditional and effective way to learn Tai Chi, but it is restricted by factors such as site and time. Although the way to learn Tai Chi by watching videos is not limited by time or place, it is not convenient for learners to observe learning from multiple perspectives since it shows sequences of actions that have been recorded from fixed viewpoints. This paper proposes a real-time interactive Tai Chi learning system based on VR and motion capture technology, in which the learner wears the head-mounted display to follow the virtual coach in the scene, and can watch the demonstration of the virtual coach in all directions; the learner can also drive and watch his own avatar, and can operate the menu by natural interaction in the virtual environment in real time through wearable motion capture device, so that he can know the effects of his training and learning of each action. The actions of the avatar and virtual coach are also compared by using a motion similarity matching algorithm, to evaluate the quality of learning and training and guide the learner to adjust his motion in real time. With the help of our system, learners can practice Tai Chi in the same way as in the real environments and drive their avatars in virtual scenes, so as to achieve effective self-learning and evaluation. [ABSTRACT FROM AUTHOR]

One-shot learning has recently attracted growing attention to produce models which can classify significant events from a few or even no labeled examples. In this paper, we introduce a deep Q-network strategy into one-shot learning (OL-DQN) to design a more intelligent learner to infer whether to label a sample automatically or request the true label for the active-learning set-up. Then we conducted experiments in the ALOI dataset for the classification of objects recorded under various imaging circumstances and a dataset for handwriting recognition composed of both characters and digits to have a performance evaluation and application analysis of the proposed model respectively, and the obtained results demonstrate that our model can achieve a better trade-off between prediction accuracy and the need of label requests compared with a purely supervised task, a prior work AOL, and a conventional active learning algorithm QBC. [ABSTRACT FROM AUTHOR]

Article presents the relevance of creating Smart Grid for public power networks and industrial enterprises. The article describes an experiment in which data were collected from current and voltage sensors from several different consumers of electric energy, using the created Smart Socket - Smart Energy device. The methods and algorithms of intellectual and spectral analysis were used to analyze and process experimental data. In this paper, we developed an algorithm for identifying the type of consumer of electric energy in the network of general power supply using the machine learning model XGBoost - extreme gradient boosted decision trees. [ABSTRACT FROM AUTHOR]

In the current age of high competition and risk in markets, technological advancements are a must for better growth and sustainability. The same applies to the agriculture industry. Every farmer has high stakes on the crops, their yield and quality. Rising water issues and need for proper methodologies for farm maintenance is a hot issue that needs to be tackled at utmost propriety. An automation of irrigation systems in farms is proposed in this research. The proposed solution is based on the Internet of Things (IoT), which would be a cheaper and more precise solution to the farm needs. A Monitoring system whose main purpose is to solve the over irrigation, soil erosion and crop-specific irrigation problem will be developed to ease and efficiently manage Irrigation problems. Since it is a well-known fact that the water is a scarce resource and over wastage of such an essential resource should be minimized. The proposed solution will be developed by establishing a distributed wireless sensor network (WSN), wherein each region of the farm would be covered by various sensor modules which will be transmitting data on a common server. Machine learning (ML) algorithms will support predictions for irrigation patterns based on crops and weather scenarios. So, a sustainable approach to irrigation is provided in this paper. [ABSTRACT FROM AUTHOR]

• Automated selection of image segmentation algorithms. • Computational cost reduction in image segmentation tasks. • Meta-learning outperformed any single algorithm explored over the whole dataset. • Machine Learning and Gradient-based image segmentation algorithms were studied. • Experiments were conducted over a widely known image dataset with 2100 examples. Image segmentation is a key issue in image processing. New image segmentation algorithms have been proposed in the last years. However, there is no optimal algorithm for every image processing task. The selection of the most suitable algorithm usually occurs by testing every possible algorithm or using knowledge from previous problems. These processes can have a high computational cost. Meta-learning has been successfully used in the machine learning research community for the recommendation of the most suitable machine learning algorithm for a new dataset. We believe that meta-learning can also be useful to select the most suitable image segmentation algorithm. This hypothesis is investigated in this paper. For such, we perform experiments with eight segmentation algorithms from two approaches using a segmentation benchmark of 300 images and 2100 augmented images. The experimental results showed that meta-learning can recommend the most suitable segmentation algorithm with more than 80% of accuracy for one group of algorithms and with 69% for the other group, overcoming the baselines used regarding recommendation and segmentation performance. [ABSTRACT FROM AUTHOR]

Business houses and marketers have been relying on social media to affect consumer opinions and purchasing behavior. In this paper a framework has been proposed to identify and quantify the emotive value of any movie trailer. The proposed framework made use of Dlib-ml (a machine learning toolkit) and a Genetic Algorithm inspired Support Vector Machine algorithm (GA i SVM) for parameter tuning and classification and emotive analysis of movie trailers. A case study comprising of 141 movies trailers released from Jan 1, 2017 till April 31, 2018 was done to investigate the relationship between emotive score of a movie trailer and financial returns associated with it. Results revealed a direct correlation between emotive score of a movie trailer and financial returns. Further, it was concluded that an emotionally intense movie trailer could result high financial returns in comparison to non-much emotionally intense trailers. • Release of movie trailer has a positive effect on the stock value of the movie. • Higher the emotional appeal of trailer, higher is the impact on stock value. • SVM performs better if its parameter tuning is done. [ABSTRACT FROM AUTHOR]

Convolutional Neural Networks (CNN) offer promising opportunities to automatically glean scientifically relevant information directly from annotated images, without needing to handcraft features for detection. Crater counting started with hand counting hundreds, thousands, or even millions of craters in order to determine the age of geological units on planetary bodies of the solar system. Automated crater detection algorithms have attempted to speed up this process. Previous research has employed computer vision techniques with handcrafted features such as light and shadow patterns, circle finding, or edge detection. This research continues, but now some researchers use techniques like convolutional neural networks that enable the algorithm to develop its own features. As the field of machine learning undergoes exponential growth in terms of paper count and research methods, the crater counting application can benefit from the new research, especially when conducting joint interdisciplinary projects. Despite these advancements, the crater counting community has not yet adopted standard methods for automating the process despite decades of research. This survey enumerates challenges for both planetary geologists and machine learning researchers, looks at the recent automatic crater detection advancements using machine learning techniques (primarily in methods using CNNs), and makes recommendations for the path toward greater automation. [ABSTRACT FROM AUTHOR]

Fuzzy inference systems with reinforcement learning are currently being used in differential games to train agents with no prior experience. However, the reinforcement learning algorithms based on actor-critic structure have a drawback that the policy is depended on a probability distribution. In this paper, a novel fuzzy deterministic policy gradient algorithm is introduced and applied to classical 1-vs-1 constant-velocity pursuit-evasion differential games. The key goal is to self-learn the optimal strategy in the continuous action domain and obtain a specific physical meaning of the fuzzy rules. The novel proposed algorithm is based on the deterministic policy gradient theorem and the agent learns the near-optimal strategy under the actor-critic structure. The fuzzy inference system is applied as approximators so that the specific physical meaning can be obtained by the linguistic fuzzy rules. Furthermore, the proposed algorithm is applied to solve the decision-making problem of pursuit-evasion differential games. The result is compared with other existing algorithms and it elucidates that the proposed algorithm outperforms the precision and convergence efficiency. [ABSTRACT FROM AUTHOR]

Multi-agent reinforcement learning (MARL) has been successfully applied in many fields. In MARL, the policy evaluation problem is one of crucial problems. In order to solve this problem, distributed Temporal-Difference (TD) learning algorithm is one of the most popular methods in a cooperative manner. Despite its empirical success, however, the theory of the adaptive variant of distributed TD learning still remain limited. To fill this gap, we propose an adaptive distributed temporal-difference algorithm (referred to as MS - ADTD) under Markovian sampling over time-varying networks. Furthermore, we rigorously analyze the convergence of MS - ADTD , the theoretical results show that the local estimation can converge linearly to the optimal neighborhood. Meanwhile, the theoretical results are verified by simulation experiments. • This paper proposes a distributed adaptive TD learning algorithm under Markovian sampling. • It theoretically analyzes the non-asymptotic convergence performance of the proposed algorithm. • It verifies the performance of the proposed algorithm by experiments. [ABSTRACT FROM AUTHOR]

Multiobjective evolutionary algorithms (MOEAs) have gained much attention due to their high effectiveness and efficiency in solving multiobjective optimization problems (MOPs). However, when solving MOPs, it is important but difficult to maintain a good balance of exploration and exploitation. In addition, some reference point based MOEAs with fixed reference points perform poorly on MOPs with irregular frontiers. Therefore, this paper proposes a new multistrategy multiobjective differential evolutionary (DE) algorithm, named RLMMDE. In RLMMDE, a multistrategy and multicrossover DE optimizer is utilized to alleviate the exploration and exploitation dilemma. An adaptive reference point activation mechanism based on RL is proposed to activate the adaptive adjustment of reference points. Moreover, a reference point adaptation method is proposed to improve the performance of RLMMDE on irregular frontier problems. Experimental results of RLMMDE tested on some benchmark test suites (i.e., ZDT, DTLZ, UF, WFG, and LSMOP) and two practical mixed-variable optimization problems show that the algorithm outperforms some advanced MOEAs. [ABSTRACT FROM AUTHOR]

Big Data is a research area where many different disciplines work together. Social media has grown in popularity as a tool for disseminating and gathering information. However, the success of social media like Twitter, Facebook, etc., has not only attracted genuine users but also spammers who utilize social graphs, famous phrases, and hashtags to spread malware. This study uses several social network analysis and visualization methods based on bibliometric data from the Web of Science to look at the structure and patterns of interdisciplinary collaborations and the latest emerging overall practice. For a better understanding of spamming behaviors on Twitter, the Twitter data set is thoroughly analyzed, and categorized into Spam and Non-Spam classifications. Earlier studies confined their scope to investigating the most negatively influential spammers by blocking the most influential spammers. However, the cumulative impact of other spammers having low individual negative influence values but higher impact values was neglected. In this article, we develop an algorithm for detecting social spam using Node Rank-based Influence Minimization (NRIM), which integrates Node Rank with the impact value of spam. The proposed spam influence minimization model also identifies spam-influential users and aids in limiting the flow of spam tweets within the Twitter network. Additionally, a detection algorithm for influential communities has been proposed to limit the spread of spam content through influential communities on the Twitter network. The primary focus of this paper is to reduce the spam impact on Twitter data by identifying influential spammers using the Node_Rank-based Influence Minimization (NRIM) algorithm. To begin, the tweets are classified into spam and non-spam using a machine learning algorithm. Furthermore, the spam observed in the Graph is analyzed, and the Spammer is passed through the NRIM algorithm to find the influential Spammers. In addition to this, the negative impact of the Spammer is reduced on the Twitter graph, and its impact is analyzed on query processing executed on Graph. The technique used for the minimization of the Spammer's negative effect on the graph reduces the query execution time by 12%. • A novel taxonomy and thorough review providing classification of spam and non spam. • Spam are discovered and the impact of influence on their spread is the main emphasis. • The spam influential value is determined using the NRIM algorithm. • The impact of the spam is reduced by Influence-based Spam reduction strategy. • Real-time statistics used in assessment, proposed method outperforms earlier method. [ABSTRACT FROM AUTHOR]

Large-scale multi-objective optimization problems (LSMOPs) bring significant challenges due to their large number of decision variables. Most of the existing algorithms fail to obtain high-quality solutions for the LSMOPs. To remedy this issue, an algorithm named dual-stage large-scale multi-objective evolutionary algorithm with dynamic learning strategy (DLMOEA-DLS) is proposed in this paper. In the DLMOEA-DLS, the entire evolution process mainly includes two stages, and each stage plays a different role in the searching process. In the first stage, the decision variables are clustering into two categories to be optimized independently for the convergence of the population. In the second stage, a dynamic learning strategy is designed to generate new offspring, in which each solution learns from a leader with better fitness and coupled control parameter for each solution is adaptively updated by learning from the historical behaviors of the solution. Moreover, an environmental selection operator is adopted to reserve promising solutions for the next iteration. To verify the performance of the DLMOEA-DLS, five state-of-the-art algorithms are used for comparison on 36 LSMOP benchmark instances, 48 LMF benchmark instances, and 6 real-world TREE benchmark instances. The experimental results demonstrate the superiority of the DLMOEA-DLS over the five state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

In this paper we propose and discuss different Deep Learning-based ensemble algorithms for a problem of low-visibility events prediction due to fog. Specifically, seven different Deep Learning (DL) architectures have been considered, from which multiple individual learners are generated. Hyperparameters of the models, including parameters concerning data preprocessing, models architecture and training procedure, are randomly selected for each model within a pre-defined discrete range. Also, every model is trained with slightly different data sampled randomly, assuring that every models introduce variety in the ensemble. Then, three different information fusion techniques are employed to build the ensemble models. The influence of the filtering process and the elitism level (the percentage of the individual models entering the ensemble) is also assessed. The performance of the proposed methodology have been tested in two real problems of low-visibility events prediction due to orographical and radiation fog, at the north of Spain. Comparison with different Machine Learning, alternative DL algorithms and meteorological-based methods show the good performance of the proposed deep learning ensembles in this problem. [ABSTRACT FROM AUTHOR]

• A novel paradigm, case learn-based, for quantitative recommendation of fault diagnosis algorithms is proposed, which is quantitative, expert-experience-free, and computation-friendly. • A three-level attribute set is defined and quantified to describe those main elements throughout the diagnosis design process. • The efficiency of the proposed paradigm is verified by an application of gearbox fault diagnosis, which also gives an example for other real PHM algorithm recommendation applications. • Applicable rules of six diagnosis algorithms are learned to realize convenient recommendations. The model has an average recommendation accuracy of 70.40% and almost no time consumption compared to real application, which shows strong capability for various real PHM algorithm design applications. • The paradigm can be extended to multi-type systems and multi-type PHM solutions. Prognostics and Health Management (PHM) is a core technology for condition-based maintenance. The diversity of PHM algorithms and the complexity of design factors make it challenging to choose an appropriate algorithm for a specific application. Consequently, automatic PHM algorithm recommendation (PHM-AR) is of great significance for PHM developers to implement rapid design of PHM systems. However, the two existing paradigms for PHM-AR rely heavily on expert experience and failure data respectively. In this paper, we propose a new third paradigm, called the case-learning-based paradigm, for quantitative recommendation of fault diagnosis algorithms, focusing on diagnosis task attribute analysis, multi-dimensional feature representation, and recommendation model construction. Specifically, to provide a quantitative basis for the recommendation of diagnosis algorithms, we define a three-level attribute set of diagnosis tasks and propose a quantitative representation method to make data structured. Then, a recommendation model based on the Classification and Regression Tree (CART) is proposed. Finally, taking gearbox as an example, we construct a diagnosis case set and verify the effectiveness of the proposed framework. Experimental results show that the average recommendation accuracy of our proposed method reaches 70.40%. These results also elucidate that the proposed method can learn the applicable rules of various diagnosis algorithms. [ABSTRACT FROM AUTHOR]

Age estimation has become inordinately significant for human beings for many reasons, such as detecting legal and criminal responsibility and other social events like a marriage license, birth certificate, etc. This paper aims to decide on the most desirable machine learning algorithm (from conventional machine learning algorithms to deep learning) for dental age estimation based on buccal bone level. The database consisted of 150 CBCT images (73 males and 77 females) from an existing base of the Faculty of Dental Medicine with Clinics, University of Sarajevo, aged 20-69. Results were obtained using the Waikato Environment for Knowledge Analysis (Weka), machine learning software in Java. Left and Right Buccal Alveolar Bone Levels are increasing with age, so they showed to be the most important attributes, especially the latter. Random Forest classifier provided the greatest result with the correlation coefficient of 0.803 and the mean absolute error of 6.022. We have also shown that considering sinus-related features can be a significant addition to the databases. Our paper is probably one of the first studies where regression algorithms based on the Support Vector Machines and Random Forest were utilized. [ABSTRACT FROM AUTHOR]

AI decisions are increasingly determining our everyday lives. At present, European anti-discrimination law is process-oriented; it prohibits the inclusion of sensitive data that is particularly protected. However, especially in the context of AI decisions, constellations can be identified in which the inclusion of sensitive characteristics will lead to better and sometimes even less discriminatory result. A result-oriented approach, therefore, might be a more fitting strategy for algorithmic decision making. In this paper we examine the legal framework for including sensitive features in a Support Vector Machine for a fictitious scenario and discuss the resulting challenges in practical application. It turns out that generally ignoring sensitive features - as has been the practice up to now - does not seem to be a fitting strategy for algorithmic decision making. A process-oriented procedure only supposedly comes closer to individual case justice: If one assumes that fewer errors occur when protected characteristics are included, individuals will ultimately also be assessed incorrectly less often, especially when one protected group is more prone to errors than the other. This paper aims to support the current debate about legal regulation of algorithmic decision making systems by discussing an often neglected perspective. [ABSTRACT FROM AUTHOR]