Search

Currently, text documents from multiple sources have become available in many fields. It is of great interest to study the relationship between documents from different sources and uncover the underlying causality. Zhu et al. (2021) proposed a joint dynamic topic model (JDTM). They classified all topics into three groups and used the "shared topics" with a fixed time lag order to characterize the shared information between two corpora. Although JDTM is a powerful tool for discovering the lead-lag relationship, there are two potential shortcomings. First, different shared topics should have distinct meanings, which should lead to different time lag orders between the two corpora. Second, for dynamic documents, not all topics are represented in each time slice, and thus topic sparsity should be considered. To address these two problems, we propose a sparse joint dynamic topic model (SJDTM) with a flexible lead-lag order. We assume a birth-and-death mechanism for all topics and a flexible lead-lag order for different shared topics. The performance of SJDTM is evaluated using both synthetic data and two real text corpora consisting of conference papers and journal papers. [ABSTRACT FROM AUTHOR]

The ranking of decision makers (DMs)' preferences for feasible states in the graph model for conflict resolution (GMCR) is crucial for accurately determining stability results. This paper addresses the issue of subjective ranking methods lacking theoretical foundation and causing ambiguity when the number of feasible states is high by proposing the implementation of the multi-attribute decision-making (MADM) method in the GMCR. The paper utilizes the average level to choose evaluation based on distance from average solution (EDAS) method for determining the DM's preference ranking, which can effectively reduce the impact of anomalous evaluations. Further, the PUL-EDAS method based on probabilistic uncertainty linguistics (PUL) is developed, which overcomes the shortcomings of the traditional EDAS method, which only applies to the simple evaluation of information. The PUL aligns with DMs' daily evaluation practice by providing an interval for the quality of qualitative linguistic evaluations. Furthermore, it utilizes an objective aggregation method to calculate comprehensive evaluation information from all DMs. In addition, the four fundamental stability definitions, applicable solely under crisp preferences, are extended to the PUL context, providing related extended definitions. Finally, to ensure the scientific validity and practicality of the proposed theory, this paper selects digital rural governance as the research context for conflict calculus analysis, comparing it with other MADM methods in the preference ranking section. [ABSTRACT FROM AUTHOR]

Condition-based maintenance (CBM) policy can avoid premature or late maintenance and reduce system failures and maintenance costs. Most existing CBM studies cannot solve the dimensional disaster problem in multi-component complex systems. Only some studies consider the constraint of maintenance resources when searching for the optimal maintenance policy, which is hard to apply to practical maintenance. This paper studies the joint optimization of the CBM policy and spare components inventory for the multi-component system in large state and action spaces. We use Markov Decision Process to model it and propose an improved deep reinforcement learning algorithm based on the stochastic policy and actor-critic framework. In this algorithm, factorization decomposes the system action into the linear combination of each component's action. The experimental results show that the algorithm proposed in this paper has better time performance and lower system cost compared with other benchmark algorithms. The training time of the former is only 28.5% and 9.12% of that of PPO and DQN algorithms, and the corresponding system cost is decreased by 17.39% and 27.95%, respectively. At the same time, our algorithm has good scalability and is suitable for solving Markov decision-making problems in large-scale state and action space. • Considering minor and major repair, we model the joint optimization of CBM and spare ordering for large multi-component systems based on MDP. • An improved DRL algorithm is presented to deal with the MDP model in large-scale discrete state and action space. • We validate our DRL algorithm has good time performance and optimal decision-making series solution via comparisons with DQN and PPO algorithm. [ABSTRACT FROM AUTHOR]

This paper aims to contribute to the extension of classical Formal Concept Analysis (FCA), allowing the management of unknown information. In a preliminary paper, we define a new kind of attribute implications to represent the knowledge from the information currently available. The whole FCA framework has to be appropriately extended to manage unknown information. This paper introduces a new logic for reasoning with this kind of implications, which belongs to the family of logics with an underlying Simplification paradigm. Specifically, we introduce a new algebra, named weak dual Heyting Algebra, that allows us to extend the Simplification logic for these new implications. To provide a solid framework, we also prove its soundness and completeness and show the advantages of the Simplification paradigm. Finally, to allow further use of this extension of FCA in applications, an algorithm for automated reasoning, which is directly built from logic, is defined. [ABSTRACT FROM AUTHOR]

Large-scale optimization problems (LSOPs) become increasingly ubiquitous but complicated in real-world scenarios. Confronted with such sophisticated optimization problems, most existing optimizers dramatically lose their effectiveness. To tackle this type of problems effectively, we propose a heterogeneous cognitive learning particle swarm optimizer (HCLPSO). Unlike most existing particle swarm optimizers (PSOs), HCLPSO partitions particles in the current swarm into two categories, namely superior particles (SP) and inferior particles (IP), based on their fitness, and then treats the two categories of particles differently. For inferior particles, this paper devises a random elite cognitive learning (RECL) strategy to update each one with a random superior particle chosen from SP. For superior particles, this paper designs a stochastic dominant cognitive learning (SDCL) strategy to evolve each one by randomly selecting one guiding exemplar from SP and then updating it only when the selected exemplar is better. With the collaboration between these two learning mechanisms, HCLPSO expectedly evolves particles effectively to explore the search space and exploit the found optimal zones appropriately to find optimal solutions to LSOPs. Furthermore, to help HCLPSO traverse the vast search space with promising compromise between intensification and diversification, this paper devises a dynamic swarm partition scheme to dynamically separate particles into the two categories. With this dynamic strategy, HCLPSO gradually switches from exploring the search space to exploiting the found optimal zones intensively. Experiments are executed on the publicly acknowledged CEC2010 and CEC2013 LSOP benchmark suites to compare HCLPSO with several state-of-the-art approaches. Experimental results reveal that HCLPSO is effective to tackle LSOPs, and attains considerably competitive or even far better optimization performance than the compared state-of-the-art large-scale methods. Furthermore, the effectiveness of each component in HCLPSO and the good scalability of HCLPSO are also experimentally verified. [ABSTRACT FROM AUTHOR]

Some cognitive and decision making experiments have demonstrated the classical decision theory may be violated. Recently, the interference effects of quantum theory have attracted a strong interest in applying some fields outside physics. It can be also used to explain the paradox in decision models. Existing some experiments and studies attribute the main reason for the existence of interference effects to uncertain information in decision process. Dempster-Shafer evidence theory extends the framework of discernment to power sets so it can describe unknown and imprecise information. This paper proposes evidential Markov decision-making model based on belief entropy to quantitatively predict and determine the value of interference effects. In new model, the frame of discernment is extended by introducing hesitant or unknown states which could be hidden by participants. Moreover, new model assumes there is no input of any information at initial states so it has the most chaotic states and is determined according to the maximum belief entropy. Finally, this paper discusses the effectiveness of new model by comparing with other methods as studying the interference effects of decision process. [ABSTRACT FROM AUTHOR]

Implementations that follow the functional programming paradigm are being used in more and more domains. As functional programming paradigm has mathematical reference transparency, refinement to functional programs contributes to improving the reliability of the transformation process and simplifying the refinement steps. However, it is a challenge to generate functional programs from specifications. Most existing transformation methods refine specifications into abstract algorithm-level programs based on loop invariants rather than functional programs. This paper proposes a novel functional program generation method based on the partition-recursion refinement rule. It establishes a novel program refinement framework based on functional theory for the first time. This is the first study to regard the whole program refinement process as a composition of abstract functions. This paper designs a recurrence-based algorithm design language (Radl+) and implements a software prototype to map Radl+ algorithms into executable Haskell programs. To prove the feasibility and efficiency of this method, this paper transforms the polynomial multiplication problem from a specification into an executable Haskell program. This case shows that compared with existing approaches, the proposed method can simplify the transformation steps and reduce the number of lines of generated code from 38 to 10. • Novel refinement framework provides a new approach to generating a functional program. • The composition of abstract functions explains the program refinement process. • Substitution rule and Recursion rule have none of the side effects. • Software prototype transforms the polynomial multiplication problem into Haskell program. [ABSTRACT FROM AUTHOR]

Forgery of a signature with the aim of deception is a serious crime. Machine learning is often employed to detect real and forged signatures. In this study, we present results which argue that robotic arms and generative models can overcome these systems and mount false-acceptance attacks. Convolutional neural networks and data augmentation strategies are tuned, producing a model of 87.12% accuracy for the verification of 2,640 human signatures. Two approaches are used to successfully attack the model with false-acceptance of forgeries. Robotic arms (Line-us and iDraw) physically copy real signatures on paper, and a conditional Generative Adversarial Network (GAN) is trained to generate signatures based on the binary class of 'genuine' and 'forged'. The 87.12% error margin is overcome by all approaches; prevalence of successful attacks is 32% for iDraw 2.0, 24% for Line-us, and 40% for the GAN. Fine-tuning with examples show that false-acceptance is preventable. We find attack success reduced by 24% for iDraw, 12% for Line-us, and 36% for the GAN. Results show exclusive behaviours between human and robotic forgers, suggesting training wholly on human forgeries can be attacked by robots, thus we argue in favour of fine-tuning systems with robotic forgeries to reduce their prevalence. • Development of a computer vision-based system for signature spoofing attack detection. • A Conditional GAN can generate "real" and "fake" signatures. • Two robots can physically replicate human signatures with pen and paper. • The GAN and both robots can fool the model and mount false-acceptance attacks. • Verification model can be defended by fine-tuning on generative and robotic forgeries. [ABSTRACT FROM AUTHOR]

Data augmentation strategies have always been important in machine learning techniques and play a unique role in model performance optimization processes. Therefore, in recent years, these techniques have become popular in the artificial intelligence field. In this paper, a new data augmentation strategy is proposed based on the interpretation algorithm of deep convolutional neural networks, i.e., constructing new training samples by deeply exploiting key features extracted from interpretable networks to achieve sample augmentation. Thus, a novel supervised data augmentation approach known as Supervised Data Augmentation–Key Feature Extraction (SDA-KFE) was proposed. By introducing the Neural Network Interpreter-Segmentation Recognition and Interpretation (NNI-SRI) algorithm, an augmentation strategy is proposed that can balance the high accuracy and high robustness of the final model while ensuring a large amount of data augmentation. The advantages of the SDA-KFE algorithm are mainly reflected in the following aspects. First, it is easy to implement. This algorithm is implemented based on the lightweight NNI-SRI algorithm, which lays the foundation for the implementation of SDA-KFE so that it can be easily implemented on convolutional neural networks. Second, this model, which is widely applicable, can be applied to almost any deep convolutional network. Through research and experiments on this proposed algorithm, SDA-KFE can be applied in graphical image binary classification and multiclassification models. Third, SDA-KFE can rapidly construct data samples with diverse variations. Under the premise of determining the classification labels of the generated samples, the distribution of the feature unit composition of the samples can be controlled. Compared with traditional data augmentation methods, SDA-KFE can control the direction of the model performance, i.e., the balance between the pursuit of high accuracy and robust performance of the model. Therefore, the novel supervised augmentation approach proposed in this paper is relevant for optimizing deep convolutional neural networks, solving model overfitting, augmenting data types, etc. The data augmentation algorithm proposed in this paper can be regarded as a useful supplement to traditional data augmentation methods, such as horizontal or vertical image flipping, cropping, color transformation, extension and rotation. [ABSTRACT FROM AUTHOR]

• EKRS is an intelligent research assistance system to recommend knowledge to scholars. • EKRS is formed through mapping two sets of IR and CRD. • IR and CRD were reconstructed based on the VSM. • LRA improving the solution process and decreasing the complexity of the UKVSM. Expert Knowledge Recommendation System (EKRS) is an intelligent research assistance system. The system is formed by mapping two sets of conceptual spaces through Institutional Repository (IR) and Core Resource Dataset (CRD) in 2018. The user knowledge pattern matching (UKPM) of EKRS has problems such as uncertain user knowledge text matching, slow update of expert knowledge, and inability to accurately track user knowledge. This paper establishes a user knowledge vector space reconstruction model (UKVSM) through the following steps to solve the above problems. Firstly, the text feature items of IR and CRD are reconstructed and the depth and density correction coefficient matrix of the original node of the text semantic meaning is calculated based on the similarity of feature items of the semantic layer. Secondly, in order to improve the efficiency of UKPM exact matching, the Lagrangian relaxation algorithm (LRA) is used to optimize the two sets of knowledge matching strategies. Finally, the real data set is extracted from the EKRS platform, and the model and algorithm proposed in this paper are tested and verified respectively, and compared with other methods. Experiments show that reconstruction model can improve the accuracy of user knowledge task assignment in EKRS, while LRA can improve the efficiency of model solving. [ABSTRACT FROM AUTHOR]

Existing hierarchical clustering (HC) algorithms generally depend on the Euclidean characteristic metric (Euclidean distance, Manhattan distance, Chebyshev distance, etc.) on Euclidean space to describe the similarity between objects, which makes the clustering process oriented to data sets with uniform and regular distribution in Euclidean space. Although such methods can visually distinguish the cluster distribution of data, it is not effective for the data sets which are densely distributed, interlaced and complex in Euclidean space. As a scalable, efficient and robust method, granular computing generally analyzes data from the perspective of similarity and proximity. In consideration of the advantages of granular computing in extracting data information from a multi-level perspective, in order to reduce the limitations of HC methods based on Euclidean features on non-Euclidean data, this paper proposes a novel HC method based on non-Euclidean feature structure. First, this paper constructs the similarity between objects based on K -dominance granularity and neighborhood search, and considers the environmental information of data points from both global and local perspectives. Secondly, a new HC method based on non-Euclidean feature structure is designed on the basis of the similarity measurement constructed in this paper. Finally, through comparative analysis, the experimental results prove that our method can more accurately identify the densely distributed and interlaced data sets in Euclidean space; it is significantly better than comparison algorithms using different Euclidean features to measure similarity; it has good robustness when additional Gaussian noise is added. [ABSTRACT FROM AUTHOR]

In fuzzy large group decision making methods, an effective clustering method can greatly reduce the complexity of decision making, and it is an important ingredient for reaching a group consensus. In this paper, a novel fuzzy large group decision making method is established using three-way clustering and an adaptive exit-delegation mechanism. Traditional clustering approaches group together individuals (isolated points) that deviate from the whole. The individuals (edge points) may exist and wander in between two or more classes. Both circumstances can lead to unstable and unreasonable clustering results. To overcome both setbacks, we propose a three-way clustering method based on the k -means clustering algorithm. The method first applies k -means clustering to perform an initial division of the universe of decision-makers. Then, in the spirit of three-way clustering, the edge points and outliers are separated from the clustering results by resorting to the three-way relationships between individuals and classes. The final clustering stems from an adaptive exit-delegation mechanism, and a consensus measure-based model determines the intra-group individual weight and inter-individual trust weight. Finally, the feasibility and effectiveness of the methodology that arises from the model designed in this paper are verified by comparative analyses. [ABSTRACT FROM AUTHOR]

This paper concerns a new incremental approach to mining data streams. It is known that patterns in a data stream may evolve over time. In many cases, we need to track and analyze the nature of these changes. In the paper, the probabilistic neural networks are considered as basic models for tracking changes in data streams. We present globally convergent stream data mining algorithms applied to problems of regression, classification, and density estimation in a time-varying (drifting) environment. The algorithms are derived from the Parzen kernel-based probabilistic neural networks working in the online mode. For each problem, a theorem is presented ensuring the L 2 convergence of the algorithm designed for tracking drifting regression, density, or discriminant functions. Illustrative examples explain in detail how to choose the bandwidth of the Parzen kernel and the learning rate of the online algorithm. The performance of all algorithms is shown in exemplary simulations. It should be noted that this paper is one of very few, in the existing literature, presenting mathematically justified stream data mining algorithms. • The incremental version of the Generalized Regression Neural Network (IGRNN) able to track drifting regression functions. • The incremental version of the Probabilistic Neural Network (IPNN) working in non-stationary environments. • Application of IPNN for tracking drifting discriminant functions. • Mathematical proofs of the L 2 convergence of all the proposed estimators. [ABSTRACT FROM AUTHOR]

Three-way decision, an essential granular computing research tool, provides an efficient solution to complex and uncertain problems. Behavioral decision theory can analyze the risk preferences of decision-makers effectively. Scholars have conducted preliminary exploration on the fusion of these two theories, but it is still challenging to describe the different types of risk preferences of decision-makers. This paper combines prospect theory with regret theory and studies the compound risk preference modeling of three-way decision to address this issue. Because three attitudes of conflicts coincide with three-way decision, many scholars have conducted multi-dimensional research on three-way conflict analysis and accomplished remarkable results. However, few relevant studies consider psychological factors and risk attitudes of decision-makers, and it is more appropriate to describe agents' attitudes on issues using linguistic terms. This paper applies the proposed three-way decision model based on compounded risk preference and probabilistic linguistic term sets to the conflict analysis problem. We utilize examples to explain the decision-making process of the proposed model and three-way conflict analysis method with the influence of the compounded risk preference under the action of reference point and regret avoidance coefficient. The illustrative example illustrates that the proposed three-way decision model can effectively solve the software development conflict analysis problem for different decision-makers and the comparative analysis shows the advantages of the proposed model and method compared with the two existing methods. Finally, we verify the performance of the three-way decision model based on compounded risk preference by UCI data sets in parameter experiments. The changes of the reference point from 10 to 0 and regret avoidance coefficient in 0, 0.15 and 0.3 respectively demonstrate the trend rule of the model's thresholds and delay-decision rate index. [ABSTRACT FROM AUTHOR]

This paper proposes a unified event-triggered sliding-mode control framework to attain the finite/fixed-time reachability of practical sliding-mode band. In event-triggered sliding-mode control, the practical sliding mode makes the size of the sliding-mode band dependent on the event function rather than the disturbance bound and sampling interval and provides better control performance due to this advantage. Under this paper's unified framework, the predesigned practical sliding-mode band can be respectively reached within a finite/fixed time by choosing different parameters. Then, different from the asymptotical convergence obtained in other investigations, the ultimate finite-time stability of the controlled system can be guaranteed. In the sliding phase, by adjusting the initial value of integration for settling time from initial value of the controlled system to the point where the sliding phase starts, a more precise estimation to settling time is obtained and can be generalized to different kinds of systems. In addition, in comparison to other results in finite-time event-triggered sliding-mode control, signum function is subtracted from the measurement error which eliminates the Zeno phenomenon and ensures the reliable operation of the digital controller in reality. Finally, a numerical example is given to verify the effectiveness of the theoretical results. [ABSTRACT FROM AUTHOR]

Interval-valued fuzzy preference relations (IVFPRs) have been applied to many real-life decision-making problems. However, most definitions of consistency of IVFPRs do not satisfy invariability to compared objects' labels. To overcome this drawback, this paper mainly focuses on the multiplicative consistency analysis of interval-valued fuzzy preference relations (IVFPRs). Firstly, this paper proposes a new multiplicative consistency of complete IVFPRs. It is proved that this new multiplicative consistency is robust and invariable to compared objects' labels. Then, the definition of acceptable incomplete IVFPRs (In-IVFPRs) is presented. To make full use of all direct and indirect evaluations of decision-makers, an algorithm is devised to evaluate the missing elements of an acceptable incomplete In-IVFPR. To comprehensively describe the closeness between any two complete IVFPRs, the total deviation of two complete IVFPRs is defined based on the p -norm of a vector. By minimizing the total deviation of two complete IVFPRs, a programming model is built to determine an interval weight vector from a complete IVFPR. Subsequently, a novel decision-making method with an In-IVFPR is proposed. Lastly, three practical and numerical examples and simulation-based comparative analyses are provided to further validate the practicability and advantages of the proposed method. [ABSTRACT FROM AUTHOR]

• This paper presents a lightweight and efficient spatial–temporal transformer network to learn collaboratively and efficiently to detect epileptic seizures. • A hierarchical FL framework is introduced to enable resource-efficient training of the detection network. • The proposed Fed-ESD mitigates the risk of a single point of failure by alleviating reliance on a centralized authority. Epilepsy is a predominant paroxysmal neurological disturbance that is usually recognized as the incidence of impulsive seizures rarely seen in medicine. Automatic detection of epileptic seizures from electroencephalogram (EEG) signals is viewed as an effective diagnosis of patients on the Internet of Medical Things (IoMT). To design a robust detection service in an IoMT environment, the EEG signals of different patients are collected from geographically distributed patients to a centralized server. However, this makes the patient's privacy prone to exposure and adds to the energy and communication costs. Also, the central server can be subject to malevolent attacks, resulting in non-efficient solutions. In this regard, for the first time, this paper presents a privacy-preserving federated learning framework for epileptic seizure detection (called Fed-ESD) from EEG signals in the fog-computing-based IoMT. A lightweight and efficient spatiotemporal transformer network is introduced to collaboratively learn spatial and temporal representations from the local data of each participant. The proposed Fed-ESD employs geographically situated fog nodes as local aggregators to enable sharing of location-based EEG signals for comparable IoMT applications. Moreover, a greedy method is introduced for deciding on the ideal fog node to be the coordinator node responsible for global aggregation during the training, thereby decreasing the reliance on the central server in the IoMT. Experimental evaluations demonstrate the efficiency of the proposed Fed-ESD in terms of detection performance, resource-efficiency, stability, and scalability for deployment in the IoMT. [ABSTRACT FROM AUTHOR]

In this paper, a new two-dimensional gaussian hierarchical priority fuzzy system (TGHPFS) is proposed to handle interval-valued data. TGHPFS first performs hierarchical clustering of the average value of interval-valued data in each dimension to generate two-dimensional gaussian membership functions of two-level rules. The two levels of rules are associated by calculating the activation strength of the second-level rules to the first-level rules and setting the connection threshold. The regularized least squares method is used to optimize the consequents of the second-level rules. The two-dimensional gaussian membership function designed in this paper is used to model the antecedents of interval-valued data, solving the correlation problem between the left and right values of interval-valued data. The effectiveness of TGHPFS is validated using real-world datasets, and the proposed method is compared with other latest methods to show the superiority of TGHPFS. [ABSTRACT FROM AUTHOR]

This paper proposes a novel algorithm twice optimization for interval type-2 fuzzy neural networks with asymmetric membership functions (TOIT2FNN-AMF), for nonlinear system identification problems. The proposed TOIT2FNN-AMF uses an asymmetric Gaussian interval type-2 membership function to enhance the network's ability to describe and solve nonlinear and uncertain problems. The twice optimization algorithm consists of structure learning and parameter learning. Firstly, this paper proposes a multi-strategy adaptive differential evolution (MSADE) algorithm as the first optimization algorithm, which is used to determine the structure and the initial values of the parameters of the TOIT2FNN-AMF. It applies the root mean square error (RMSE) of the TOIT2FNN-AMF as the fitness function to determine the structure (number of rules) and initial parameters of the IT2FNN by searching for the RMSE values under different structures. When the fitness value reaches the minimum, that is, the RMSE value of the TOIT2FNN-AMF, the corresponding number will become the optimal one of fuzzy rules of the TOIT2FNN-AMF. Then, the second optimization algorithm of the TOIT2FNN-AMF turns into a hybrid optimization algorithm composed of an adaptive moment estimation (Adam) algorithm and recursive least squares (RLS) algorithm. Adam is used to optimize the antecedent parameters of TOIT2FNN-AMF rules, so as to maintain rapid convergence without generating oscillation during the training process; RLS is used to optimize the consequent parameters of TOIT2FNN-AMF rules, so that the network parameters can be optimized rapidly. In this way, the problems of excessive parameters to be adjusted and excessive slow convergence of the network can be solved. Finally, this paper evaluates the proposed TOIT2FNN-AMF by testing on problems of nonlinear system identification and chaotic time-series prediction. The simulation results are compared with those of similar methods in the existing literatures, which demonstrates that the proposed TOIT2FNN-AMF model yields a lower RMSE value and a simpler network structure than the other type-2 fuzzy neural networks (T2FNNs). [ABSTRACT FROM AUTHOR]

Federated Continual Learning (FCL) approaches exist two major problems of the probability bias and the imbalance in parameter variations. These two problems lead to catastrophic forgetting of the network in the FCL process. Therefore, this paper proposes a novel FCL framework, Federated Probability Memory Recall (FedPMR), to mitigate the probability bias problem and the imbalance in parameter variations. Firstly, for the probability bias problem, this paper designs the Probability Distribution Alignment (PDA) module, which consolidates the memory of old probability experience. Specifically, PDA maintains a replay buffer and uses the probability memory stored in the buffer to correct the offset probabilities of the previous tasks during the two-stage training. Secondly, to alleviate the imbalance in parameter variations, this paper designs the Parameter Consistency Constraint (PCC) module, which constrains the magnitude of neural weight changes for previous tasks. Concretely, PCC applies a set of adaptive weights to subsets of the regularization term that constrains parameter changes, forcing the current model to be sufficiently close to the past model in parameter space distance. Experiments with various levels of task similitude across clients demonstrate that our technique establishes the new state-of-the-art performance when compared to previous FCL approaches. [ABSTRACT FROM AUTHOR]

• The finish time and total costs are extracted as the optimization objectives based on problem characteristics. • The dynamic task disassembling mechanism (DTDM) and four technician scheduling modes are novelly proposed. • A flexible time is introduced into DTDM to provide decision-makers with a larger pool of choices to meet specific needs. • A new solution method suitable for both AMTS model and ATMS-DTDM model is designed based on PSO and MOPSO algorithms. This paper focuses on the aviation maintenance technician scheduling (AMTS) problem and formulates AMTS and AMTS-DTDM model with a practical dynamic task disassembly mechanism (DTDM) and arrange maintenance technicians across shifts in short-term maintenance situations (less than 24–48 h). In DTDM, four technician scheduling modes are devised to flexibly disassemble the overtime work or reassigned it to other maintenance technicians according to work efficiency and progress, which could shorten the maintenance time and save the total cost. Moreover, a flexible time interval is designed to adjust the boundary of task disassembly. To further study the effectiveness of DTDM in reducing maintenance time and total costs, this paper divides both the AMTS and AMTS-DTDM models into three sub-models (i.e., two single-objective models and one multi-objective model), respectively. After that, we design solution methods as well as the encoding schemes compatible with both the AMTS model and the AMTS-DTDM model for different problem scales. Finally, to verify the effectiveness of DTDM, four groups of experiments are set up and particle swarm optimization (PSO) and multi-objective particle swarm optimization (MOPSO) are applied to compare AMTS and AMTS-DTDM sub-models. The experimental results show that the AMTS-DTDM model can effectively shorten the maintenance time and reduce the total costs for different scaled problems. Furthermore, the flexible time interval can facilitate more options for airlines to adjust the two above objectives to a small extent. The increase in flexible time may weaken the advantages of the AMTS-DTDM model while improving the rationality of the ultimate technician scheduling scheme. [ABSTRACT FROM AUTHOR]

Stochastic configuration networks (SCNs) have universal approximation capability and fast modeling properties, which have been successfully employed in large-scale data analytics. Based on SCNs, Stochastic configuration networks with block increments (BSC) use the node block increments mechanism to improve training speed but increase the complexity of the model. This paper presents a parallel configuration method (PCM), develops an extension of the original BSC with chaos theory and proposes stochastic configuration networks with chaotic maps (SCNCM), and establishes a hierarchical learning strategy (HLS) to enhance the compactness and construction speed of the model. Firstly, PCM randomly assigns the input weights w and biases b of hidden layer nodes by using uniform and normal distributions. In PCM, an iterative learning algorithm is intended to generate the scope control set and improve configuration efficiency. Secondly, the paper presents two kinds of stochastic configuration networks with chaotic maps, which are SCNCM-I and SCNCM-II. SCNCM-I adjusts block size by using multiple error values and chaotic maps to improve the training speed. Based on SCNCM-I, SCNCM-II utilizes node removal mechanism to enhance the compactness. Finally, HLS integrates with SCNCM-I, SCNCM-II, and the Harris-hawks optimization algorithm (HHO). The purpose of training is to enhance the training speed and compactness for three algorithms. The experiments are conducted on four benchmark data sets and an industrial application shows its effectiveness. [ABSTRACT FROM AUTHOR]

Many real-life problems can be formulated as numerical optimization problems. Such problems pose a challenge for researchers when designing efficient techniques that are capable of finding the desired solution without suffering from premature convergence. This paper proposes a novel evolutionary algorithm that blends the exploitative and explorative merits of two main evolutionary algorithms, namely the Stochastic Fractal Search (SFS) and a Differential Evolution (DE) variant. This amalgam has an effective interaction and cooperation of an ensemble of diverse strategies to derive a single framework called Iterative Cyclic Tri-strategy with adaptive Differential Stochastic Fractal Evolutionary Algorithm (Ic3-aDSF-EA). The component algorithms cooperate and compete to enhance the quality of the generated solutions and complement each other. The iterative cycles in the proposed algorithm consist of three consecutive phases. The main idea behind the cyclic nature of Ic3-aDSF-EA is to gradually emphasize the work of the best-performing algorithm without ignoring the effects of the other inferior algorithm during the search process. The cooperation of component algorithms takes place at the end of each cycle for information sharing and the quality of solutions for the next cycle. The algorithm's performance is evaluated on 43 problems from three different benchmark suites. The paper also investigates the application to a set of real-life problems. The overall results show that the proposed Ic3-aDSF-EA has a propitious performance and a reliable scalability behavior compared to other state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

This paper addresses the bearing-only distributed localization, which is more accurate and more reliable than traditional localization technologies. A new bearing-only distributed localization framework is proposed to solve localization problems for both static sensor networks and mobile multi-agent systems (MASs). Based on the complex coordinate representation, a novel bearing-only distributed localization algorithm for sensor networks is proposed. The algorithm combines the orientation estimation and the position estimation to solve the orientation alignment problem, which makes the compasses no longer needed for the localized networks. The localization for the mobile MASs is also studied and the corresponding localization algorithm is designed, which is more general and more challenging. The key to obtaining positions of moving agents, is the velocity estimator in the proposed localization algorithm which makes the estimated positions and velocities converge to the true value simultaneously. A distinctive advantage of the localization framework proposed in this paper is that the bearing-only localization algorithms can be applied to more general systems, such as sensor networks and MASs. Simulations and experiment are presented to verify the proposed algorithms. [ABSTRACT FROM AUTHOR]

Nowadays, cybersecurity challenges and their ever-growing complexity are the main concerns for various information technology-driven organizations and companies. Although several intrusion detection systems have been introduced in an attempt to deal with zero-day cybersecurity attacks, computer systems are still highly vulnerable to various types of distributed denial of service (DDoS) attacks. This complicated cyber-attack caused many system failures and service disruptions, resulting in billions of dollars of financial loss and irrecoverable reputation damage in recent years. Considering the nonnegligible importance of business continuity in the Industry 4.0 era, this paper presents a comprehensive, systematic survey of DDoS attacks. It also proposes a hierarchy for this severe cyber threat, besides conducting deep comparisons from various perspectives between the studies published by reputed venues in this area. Furthermore, this paper recommends the most effective defensive strategies, with a focus on recently offered fuzzy-based detection methods, to mitigate such threats and bridge the gaps existing in the current intrusion detection systems and related works. The outcomes and key findings of this survey paper are highly advantageous for private companies, enterprises, and government agencies to be implemented in their local or global businesses to significantly improve business sustainability. [ABSTRACT FROM AUTHOR]

Compared with single-task learning, multi-tasks can obtain better classifiers by the information provided by each task. In the process of multi-task data collection, we always focus on the target task data in the training process, and ignore the non-target task data and unlabeled data that may be contained in the target task. In response to this issue, this paper introduces auxiliary or Universum into semi-supervised multi-task problem, and proposes a multi-task support vector machine (SU-MTLSVM) method based on semi-supervised learning to handle the case where each task contains the labeled, unlabeled, and Universum samples in the training set. This method introduces Universum as prior knowledge and provides high-dimensional information for semi-supervised learning, and builds a unique classifier from a large amount of unlabeled data. We then use KKT conditions and Lagrangian method to optimize the formulation of the model, and get the model parameters. Finally, we collect different data sets in the experiment part, and compare the performance of multiple baselines with the proposed method. Experiments prove that the method proposed in this paper is more effective for multi-task applications. [ABSTRACT FROM AUTHOR]

In this paper, a novel adaptive generalized Nash equilibrium (GNE) seeking algorithm is designed, in order to address the non-cooperative game with private inequality constraints under high-dimensional input dead-zone. That is to say, the dead-zone dynamics may be thought of as a generic high-dimensional convex set, and the introduction of two methods distinguishes our works in seeking the GNE of non-cooperative games. On the one hand, a two-time-scale structure based on singular perturbation method is led into the design of GNE seeking algorithm, where the fast dynamics part rapidly eliminates the influence of input dead-zone, and the slow dynamics part drives the players' action to the GNE. On the other hand, adaptive penalty method is utilized to ensure the player's action enters the inequality constraints set without a prior estimation of centralized information for penalty parameters. The algorithm in this paper realizes complete distribution and parameter independence, making it easy to apply in practical programming. At last, several numerical examples regarding the electricity markets are employed to verify the effectiveness of the theoretical results. [ABSTRACT FROM AUTHOR]

Surrogate-assisted multi-objective evolutionary algorithms have been paid much attention to solve expensive multi-objective problems in recent years. However, with the number of objectives increasing, an improper solution may be picked for expensive objective evaluation due to the accumulation error of approximated values on objective functions. Furthermore, the time to construct surrogate models for all objectives will significantly increase. Thus, in this paper, Gaussian process (GP) models are proposed for performance indicators instead of for objective functions. Furthermore, solutions are selected from either of two ways to be evaluated using the expensive objective function. When there are non-dominated solutions found so far that are approximated, they will be exactly evaluated using the objective function. Otherwise, the solution with the maximum approximation uncertainty among the current population will be evaluated using the real objective functions. The efficiency of the presented approach is validated on the DTLZ test suite with 3, 6, 10, 15, and 20 objectives, MaF benchmark problems with 3, 6, 10, 15, and 20 objectives, and a real-world optimization problem called filter design. The experimental results show that the method proposed in this paper is competitive compared to recently proposed peer algorithms for expensive many-objective problems. [ABSTRACT FROM AUTHOR]

In this paper, we present a connectivity-preserving performance function approach for the distributed output-feedback synchronized tracking of uncertain heterogeneous nonlinear multiagent systems in a nonstrict-feedback form. Compared with existing output-feedback cooperative control results using neural networks, this paper contributes to developing a universal output-feedback synchronized control methodology that uses a connectivity-preserving performance function to ensure both initial network connectivity and preselected synchronization performance with a designable convergence time. To this end, a neural-network-based adaptive observer for each follower is designed to ensure the boundedness of estimation errors of unmeasurable states. Then, local event-triggered synchronized trackers using only relative output information and the connectivity-preserving performance function are constructed to guarantee the closed-loop stability in a low-complexity sense, where no adaptive neural networks and command filters are not required in the local trackers. Finally, a purely academic example and a practical platoon-control problem of multiple uncertain vehicular systems are considered to clarify and verify the proposed connectivity-preserving performance function approach in the simulation. [ABSTRACT FROM AUTHOR]

• The paper is the first batch that tries to inquiry the GPS of the PDIMNNs. Furthermore, unlike the previous literature [11,21,35] , the nonlinear substitution that converts the PDs system to a constant delays system is not used. Because the subsequent pro-cessing of the converted system is complicated, this paper directly performs on the original PDs system. • The paper employs the non-reduced-order method, which r-efrains the double-dimensional problem after reduced-order [3–5,23,24]. In practical applications, the method is more ponderable and si-gnificative for second-order scheme under controller. • In the paper, devise both the feedback controller and the adaptive controller for the first time to achieve the GPS of the PD-IMNNs. The advantages of the two types of controllers are compar-ed through numerical examples and simulations. The mathematical model is closer to reality, and selecting the appropriate controller in application can further reduce control expenses. This article probes into the global polynomial stabilization (GPS) of proportional delayed inertial memristive neural networks (PDIMNNs). Here, ruling out the reduced-order way, discuss the GPS of PDIMNNs under the second-order scheme directly. Firstly, a feedback controller is designed to make the system self-stabilizing. By designing suitable Lyapunov functional with adjustable parameters and combining with inequality techniques, two algebraic criteria are obtained to realize the GPS of the PDIMNNs. Owing to the conservatism caused by the ineluctable inequality scaling, it is worth noting that the controller gains are greater than the actual requirements. To further save control expenses, employing an adaptive controller to make the system stabilized. Finally, three numerical examples which sustain the usability of the obtained theoretical conclusions are shown. [ABSTRACT FROM AUTHOR]

• Algorithmic trading using self-attention based recurrent reinforcement learning is developed. • Self-attention layer reallocates temporal weights in the sequence of temporal embedding. • Hybrid loss feature is incorporated to have predictive and reconstructive power. Algorithmic trading based on machine learning has the advantage of using intrinsic features and embedded causality in complex stock price time series. We propose a novel algorithmic trading model based on recurrent reinforcement learning, optimized for making consecutive trading signals. This paper elaborates on how temporal features from complex observation are optimally extracted to maximize the expected rewards of the reinforcement learning model. Our model incorporates the hybrid learning loss to allow sequences of hidden features for reinforcement learning to contain the original state's characteristics fully. The self-attention mechanism is also introduced to our model for learning the temporal importance of the hidden representation series, which helps the reinforcement learning model to be aware of temporal dependence for its decision-making. In this paper, we verify the effectiveness of proposed model using some major market indices and the representative stocks in each sector of S&P500. The augmented structure that we propose has a significant dominance on trading performance. Our proposed model, self-attention based deep direct recurrent reinforcement learning with hybrid loss (SA-DDR-HL), shows superior performance over well-known baseline benchmark models, including machine learning and time series models. [ABSTRACT FROM AUTHOR]

Negation is an important operation in evidence theory, whose idea is to consider the opposite of events, can deal with some problems with uncertainties from the opposite side and obtain information behind probability distribution. In classical D-S theory (Dempster-Shafer's theory), there are already many negation methods existed on real number field and many properties of which have been discovered. However, in complex evidence theory, which based on complex number field, negation is still an open problem. In order to deal with some problems like those in D-S theory, a new negation method for CBBA (Complex Basic Belief Assignment) should be proposed. In this paper, a new negation method called CBBA exponential negation will be presented, which can be seen as a generalization from BBA (Basic Belief Assignment) to CBBA. This proposed negation transforms a CBBA to another one with the entropy increased simultaneously. Also, some properties of this negation will be discussed such as invariance, convergence, fixed point, distribution of Pascal triangle, convergence speed, impact on negation convergence and so on. Besides, most among them will be strictly proved in this paper. Furthermore, a new entropy for CBBA and some numerical examples will be presented, and we will study the proposed negation from the view of entropy. Finally, an application of CBBA exponential negation will be shown in the end. [ABSTRACT FROM AUTHOR]

Telemedicine provides remote online services for digital diagnosis and treatment via the Internet. However, there is a risk of data leakage during transmission. Therefore, data protection is an important challenge for telemedicine. Chaos is widely used in image, audio, and EEG encryption because of its unique characteristics of unpredictability, nonlinearity, and sensitivity to an initial state. However, some chaotic maps have various security issues. To solve these problems, this paper proposes a K-sine-transform-based coupled chaotic system (K-STBCCS), combining any two one-dimensional chaotic mappings to generate a new chaos mapping. To demonstrate the reliability of the system, this paper generates three new chaotic mappings using K-STBCCS and analyzes their performance. Using the chaotic mapping generated by K-STBCCS, this paper further proposes an EEG signal encryption scheme based on the confusion-diffusion principle. The purpose of confusion is to separate adjacent EEG signals, while the purpose of diffusion is to change the value of EEG signals. Among them, the diffusion operation uses positive and negative diffusion to reduce the correlation between the ciphertext and the original signals. The experimental results and security analysis show that the proposed EEG signal encryption scheme performs well and passes the rigorous cryptographic security test. [ABSTRACT FROM AUTHOR]