Search

This paper presents a new hybrid genetic-particle swarm optimization (GPSO) algorithm for solving multi-constrained optimization problems. This algorithm is different from the traditional GPSO algorithm, which adopts genetic algorithm (GA) and particle swarm optimization (PSO) in series, and it combines PSO and GA through parallel architecture, so as to make full use of the high efficiency of PSO and the global optimization ability of GA. The algorithm takes PSO as the main body and runs PSO at the initial stage of optimization, while GA does not participate in operation. When the global best value (gbest) does not change for successive generations, it is assumed that it falls into local optimum. At this time, GA is used to replace PSO for particle selection, crossover and mutation operations to update particles and help particles jump out of local optimum. In addition, the GPSO adopts adaptive inertia weight, adaptive mutation parameters and multi-point crossover operation between particles and personal best value (pbest) to improve the optimization ability of the algorithm. Finally, this paper uses a nonlinear constraint problem (Himmelblau's nonlinear optimization problem) and three structural optimization problems (pressure vessel design problem, the welded beam design problem and the gear train design problem) as test functions and compares the proposed GPSO with the traditional GPSO, dingo optimization algorithm, whale optimization algorithm and grey wolf optimizer. The performance evaluation of the proposed algorithm is carried out by using the evaluation indexes such as best value, mean value, median value, worst value, standard deviation, operation time and convergence speed. The comparison results show that the proposed GPSO has obvious advantages in finding the optimal value, convergence speed and time overhead. [ABSTRACT FROM AUTHOR]

This review aims to exploit a study on different benchmark test functions used to evaluate the performance of Meta-Heuristic (MH) optimization techniques. The performance of the MH optimization techniques is evaluated with the different sets of mathematical benchmark test functions and various real-world engineering design problems. These benchmark test functions can help to identify the strengths and weaknesses of newly proposed MH optimization techniques. This review paper presents 215 mathematical test functions, including mathematical equations, characteristics, search space and global minima of the objective function and 57 real-world engineering design problems, including mathematical equations, constraints, and boundary conditions of the objective functions carried out from the literature. The MATLAB code references for mathematical benchmark test functions and real-world design problems, including the Congress of Evolutionary Computation (CEC) and Genetic and Evolutionary Computation Conference (GECCO) test suite, are presented in this paper. Also, the winners of CEC are highlighted with their reference papers. This paper also comprehensively reviews the literature related to benchmark test functions and real-world engineering design challenges using a bibliometric approach. This bibliometric analysis aims to analyze the number of publications, prolific authors, academic institutions, and country contributions to assess the field's growth and development. This paper will inspire researchers to innovate effective approaches for handling inequality and equality constraints. [ABSTRACT FROM AUTHOR]

This paper presents an improved arithmetic optimization algorithm that incorporates hybrid elite pool strategies to address the limitations of the arithmetic optimization algorithm (AOA). In AOA, the linear mathematical optimization acceleration (MOA) function cannot balance global exploitation and local exploration well. Therefore, the accuracy and convergence speed of the algorithm cannot be guaranteed. To improve the performance of AOA, this paper reconstructed a nonlinear MOA function, which is expected to balance the exploitation and the exploration of AOA. Furthermore, four hybrid elite pool strategies are integrated to enhance the ability to escape local optima. The proposed algorithm inherits the fast convergence of AOA and develops the performance of escaping local optima. Numerical experiment results on benchmark functions and engineering problems show that the proposed algorithm outperforms other compared meta-heuristic algorithms in terms of convergence speed and accuracy. [ABSTRACT FROM AUTHOR]

In this paper, a new bi-objective hub facility location problem is studied where the demands of customers are stochastic and follow a normal distribution function. The first objective is to minimize total amount of value at risk which has a new probabilistic criterion and optimizes amount of lost demand. In addition, the second objective minimizes total costs of the network. Furthermore, some constraints are nonlinear in this model and it is aimed to linearize them by efficient approximate procedures. This paper tries to apply three efficient solution procedures to solve the bi-objective model. Thus, three algorithms are $$\varepsilon $$ -constraint, non-dominated sorting genetic algorithm-II (NSGA-II), and multi-objective particle swarm optimizers (MOPSO) which are creativity ways used in this paper to approximate the Pareto-optimal solutions. Taguchi experimental design is used to find the right parameter settings for metaheuristic algorithms. A comparative study of three proposed algorithms demonstrates the most effectiveness algorithm with respect to five existing performance measures for numerous test problems. Finally, the comparison within each two algorithms is completed by applying multiple statistical tests and diagrams. The obtained solutions by MOPSO are better than NSGA-II at 95 % confidence level. [ABSTRACT FROM AUTHOR]

Archimedes optimization algorithm (AOA) is a new meta-heuristic algorithm which is based on Archimedes principle and mimics the buoyancy force received by an object in water. The AOA is designed according to physical principles and has been the object of many scholars' research because of its simple and reliable performance. In the course of the study, this paper finds that the AOA is flawed. In the iterative update of the algorithm, the buoyancy principle applied to the object is not completely followed. Through the investigation and analysis of this problem, it is found that the algorithm design which follows the buoyancy principle completely is more advantageously and persuasively, and named the corrected algorithm CAOA. The performance of the CAOA and other comparison optimization algorithms is tested in benchmark functions CEC2017 under equal conditions to verify the ideas proposed in this paper. In the solution accuracy with dimensions of 30 and 50, the comprehensive score of the CAOA is 31 and 33 and ranks first in all algorithms. In the statistical analysis, the CAOA compared with other algorithms one by one, and achieved the best results in all test functions. When compared with other algorithms, the CAOA ranked first. It is hoped that the verification of the ideas in this paper will help the AOA to develop better and optimize the development of the algorithm. [ABSTRACT FROM AUTHOR]

Complex system optimization is an emerging research topic in the field of evolutionary computation, whose goal is to handle complex systems with multiple coupled subsystems, each including multiple objectives and multiple constraints in real-world applications. This paper proposes a multi-system genetic algorithm (MSGA), stemming from implicit parallelism in population-based search algorithms, to solve multiple coupled subsystems simultaneously in a complex system. The proposed MSGA is composed of within-subsystem evolution and cross-subsystem migration operators. The objective of the former is to optimize each subsystem by appropriate search strategies, and the objective of the latter is to exchange information between multiple subsystems by migration, which is based on the similarity probability of objectives and constraints, and the intersection probability of solutions in different subsystems. During migration across subsystems, three statistical approaches of measuring similarity and three metrics of solution intersection in information theory are used to calculate these probabilities. Performance is tested on a set of multi-subsystem benchmark functions, and the simulation results show that cross-subsystem migration plays the key role for the performance of MSGA. Furthermore, the proposed MSGA is compared with other competitive algorithms, and results show that it is a promising multi-system optimization algorithm. In summary, the contribution of this paper is the introduction of multi-system optimization to the EA community. [ABSTRACT FROM AUTHOR]

Human facial analysis (HFA) has recently become an attractive topic for computer vision research due to technological progress and mobile applications. HFA explores several issues as gender recognition (GR), facial expression, age, and race recognition for automatically understanding social life. This study explores HFA from the angle of recognizing a person's gender from their face. Several hard challenges are provoked, such as illumination, occlusion, facial emotions, quality, and angle of capture by cameras, making gender recognition more difficult for machines. The Archimedes optimization algorithm (AOA) was recently designed as a metaheuristic-based population optimization method, inspired by the Archimedes theory's physical notion. Compared to other swarm algorithms in the realm of optimization, this method promotes a good balance between exploration and exploitation. The convergence area is increased By incorporating extra data into the solution, such as volume and density. Because of the preceding benefits of AOA and the fact that it has not been used to choose the best area of the face, we propose utilizing a wrapper feature selection technique, which is a real motivation in the field of computer vision and machine learning. The paper's primary purpose is to automatically determine the optimal face area using AOA to recognize the gender of a human person categorized by two classes (Men and women). In this paper, the facial image is divided into several subregions (blocks), where each area provides a vector of characteristics using one method from handcrafted techniques as the local binary pattern (LBP), histogram-oriented gradient (HOG), or gray-level co-occurrence matrix (GLCM). Two experiments assess the proposed method (AOA): The first employs two benchmarking datasets: the Georgia Tech Face dataset (GT) and the Brazilian FEI dataset. The second experiment represents a more challenging large dataset that uses Gallagher's uncontrolled dataset. The experimental results show the good performance of AOA compared to other recent and competitive optimizers for all datasets. In terms of accuracy, the AOA-based LBP outperforms the state-of-the-art deep convolutional neural network (CNN) with 96.08% for the Gallagher's dataset. [ABSTRACT FROM AUTHOR]

Most production scheduling problems, including standard flexible job shop scheduling problems, assume that machines are continuously available. However, in most cases, due to preventive maintenance activities, machines may not be available for a certain time. Meanwhile, in the entire workshop production process, the transportation process of workpieces cannot be ignored. Therefore, the impact of transportation on the production planning should be considered in the scheduling process. To consider both preventive maintenance and transportation processes in the flexible job shop scheduling problem, this paper proposes a flexible job shop scheduling problem considering preventive maintenance activities and transportation processes and establishes a multi-objective flexible job shop scheduling model optimizing the total energy consumption and total makespan. Furthermore, a multi-region division sampling strategy-based multi-objective optimization algorithm integrated with a genetic algorithm and a differential evolution algorithm (MDSS-MOGA-DE) is proposed to solve the model. In the proposed algorithm, a multi-region division sampling strategy and two evaluation functions are utilized to improve the diversity of solutions. In addition, this paper combines a genetic operation and a differential operation to further enhance the search ability of the algorithm. The validity of the algorithm is verified by a real case. The computational results reveal that the proposed model and algorithm obtain appropriate results and have the potential to be applied to other similar problems. [ABSTRACT FROM AUTHOR]

This paper analyzes the numerical stability of Markowitz portfolio optimization model, by identifying and studying a source of instability, that strictly depends on the mathematical structure of the optimization problem and its constraints. As a consequence, it is shown how standard portfolio optimization models can result in an unstable model also when the covariance matrix is well conditioned and the objective function is numerically stable. This depends on the fact that the linear equality constraints of the model very often suffer of almost collinearity and/or bad scaling. A theoretical approach is proposed that exploiting an equivalent formulation of the original optimization problem considerably reduces such structural component of instability. The effectiveness of the proposal is empirically certified through applications on real financial data when numerical optimization approaches are needed to compute the optimal portfolio. Gurobi and MATLAB's solvers quadprog and fmincon are compared in terms of convergence performances. [ABSTRACT FROM AUTHOR]

In this paper Enhanced Heat Transfer Search algorithm and Enriched Replicated Coronary Circulation System Optimization algorithm are applied to solve the real power loss reduction problem. Key objectives of the paper are Voltage stability enhancement, Voltage deviation minimization and real power loss reduction. Heat transfer algorithm has been designed on the basis of law of thermodynamic state. It is regular procedure that any organism will try to attain to reach the state of equilibrium with respect to the environments. Heat transfer will occur by conduction, convection and radiation. Through arbitrarily engendered population, Heat transfer algorithm is designed with molecules and the temperature conditions. Population is modernized in every generation through arbitrarily designated heat transmission methods. In the enhanced heat transfer search algorithm reinforcement of population has been intermingled with the algorithm. Synchronized heat transfer and population reinforcement are amalgamated into the Heat Transfer Search algorithm to augment the quickness in the exploration procedure, to progress the convergence degree, and to evade local optimum. Replicated Coronary Circulation System Optimization Algorithm has been designed by emulating the activities of human heart veins (coronary artery progress). In the proposed algorithm contender solution is formulated by considering capillaries actions. Subsequently the Coronary progress factor evaluates the solution and population the area has been originated capriciously. In the Enhanced Replicated Coronary Circulation System Optimization Algorithm the Heart memory will stock pile the finest solutions and modernize them sequentially. The HMYempowers to have a sturdier exploitation and improved convergence. Additionally, method applied to modify some modules of capillaries frontrunners which helps the procedure to spurt from local optima solution. Proposed Enhanced Heat Transfer Search algorithm and Enriched Replicated Coronary Circulation System Optimization algorithm are appraised in IEEE 30 bus system and IEEE 14, 30, 57, 118, 300 bus test systems without considering the voltage constancy index. True power loss lessening, voltage divergence curtailing, and voltage constancy index augmentation have been attained. [ABSTRACT FROM AUTHOR]

In the literature, loop fusion is an effective optimization technique which tries to enhance parallelizing compilers' performance via memory hierarchy management, and all its competing criteria create an NP-hard problem. This paper proposes an evolutionary algorithm that aims to achieve a profitable loop order which maximizes fusion taking into account register size, parallelism and data reuse advancement. Besides, this method preserves prerequisite relations between the loops by encoding each distinct loop sequence as the shortest common supersequence (SCS) of the related dependence graph. Regarding the related optimization methods that only focus on fusion, this set of metrics, an evolutionary algorithm and also the shortest common supersequence problem have not been considered before in this area. Despite all the envisaged complexities, experimental results confirm the accuracy and advantage of the proposed approach. But due to evolutionary methods effect on raising the compilation time, the proposed algorithm is only applicable when this issue is not prominent, in comparison with the quality of the outcome. [ABSTRACT FROM AUTHOR]

A memetic algorithm based on MOEA/D is presented to deal with the uncapacitated multiobjective examination timetabling problem in this paper. The examination timetabling problem is considered as a two-objective optimization problem in this paper, while it is modeled as a single-objective optimization problem generally. The framework of a multiobjective evolutionary algorithm with decomposition (MOEA/D) is first employed to guide the evolutionary process. Two special local search operators are designed to find better individuals. The proposed algorithm is tested on 11 benchmark examination timetabling instances. Experimental results prove that the proposed algorithm can produce a promising set of nondominated solutions for each examination timetabling instance. [ABSTRACT FROM AUTHOR]

In this paper, Belleville Washer Search optimization algorithm, Papilionoidea Optimization Algorithm, Opposition-based Learning integrated with Papilionoidea Optimization Algorithm, Chaotic Local Search integrated with Papilionoidea Optimization Algorithm and then both the Opposition-based Learning and Chaotic Local Search combined with Papilionoidea Optimization Algorithm are designed to solve the optimization problem. Key objectives of the paper are the minimization of voltage deviation, voltage stability enhancement and power loss reduction. Belleville washer is coned-disc spring, which can encumber the length of its axis by static and dynamic mode. Belleville washers are helpful for modification since dissimilar breadth can be exchanged within and away and it can be used to attain vast tune of coil rate. They are perfect in circumstances wherever a weighty coil power is necessary with least complimentary distance end to end and firmness previous to attainment of firm elevation. Fundamental law of force has been employed to design the Belleville Washer Search optimization algorithm. Belleville washer impresses a force on the particle in horizontal direction and an external force pressure the particle in anti-clockwise direction. Belleville washer's force is equal to the peripheral force in the period of time the particle will be in balanced condition. Initially, the particle moves certain distance, and Belleville washer's force will tender a resistance over the particle during the time peripheral force acting over the particle. Stiffness property of the Belleville washer is tailored relative to the objective function. Force and movement of the particle have been observed. Papilionoidea Optimization Algorithm is based on performance of the Papilionoidea in nutrition foraging. Opposition-based Learning is a technique integrated with Papilionoidea Optimization Algorithm. Through this, exploration capability of the process will be enriched and convergence rate will be augmented. Then, in this work, Chaotic Local Search integrated with Papilionoidea Optimization Algorithm. Through this integration, the algorithm will balance the exploration and exploitation equally. Then, both the Opposition-based Learning and Chaotic Local Search has been integrated with Papilionoidea Optimization Algorithm. With and without voltage stability, index Belleville Washer Search optimization algorithm, Papilionoidea Optimization Algorithm, Opposition-based Learning integrated with Papilionoidea Optimization Algorithm, Chaotic Local Search integrated with Papilionoidea Optimization Algorithm and then both the Opposition-based Learning and Chaotic Local Search combined with Papilionoidea Optimization Algorithm are corroborated in standard IEEE 30-bus test systems. Proposed algorithms condensed the power loss competently with augmentation in voltage stability and minimization of voltage deviation. [ABSTRACT FROM AUTHOR]

This paper proposes a new population-based optimization algorithm called hunter–prey optimizer (HPO). This algorithm is inspired by the behavior of predator animals such as lions, leopards and wolves, and preys such as stag and gazelle. There are many scenarios of animal hunting behavior, and some of them have transformed into optimization algorithms. The scenario used in this paper is different from the scenario of the previous algorithms. In the proposed approach, a prey and predator population, and a predator attacks a prey that moves away from the prey population. The hunter adjusts his position toward this far prey, and the prey adjusts his position toward a safe place. The search agent's position that was the best value of the fitness function considered a safe place. The HPO algorithm implemented on several test functions to evaluate its performance. Also, to performance verification, the proposed algorithm is applied to several engineering problems. The results showed that the proposed algorithm performed effective in solving test functions and engineering problems. [ABSTRACT FROM AUTHOR]

In this paper, we proposed a data relay system consisting of multiple unmanned aerial vehicles (UAVs) and acess point (AP) with the aim of maximizing the minimum receiving rate of Internet of things devices (IoTDs) as the optimization objective. To enhance the utilization of communication resources, we employed non-orthogonal multiple access (NOMA) and successive interference cancellation (SIC) techniques at the AP side. In addition, we used the Wireless power transmission (WPT) technique on the UAV side to charge the IoTDs and prolong their operational lifetime. We formulated the problem as a mixed integer nonlinear programming (MINLP). To solve the MINLP problem, it was decomposed into four subproblems: connection scheduling between UAVs and IoTDs, time slot allocation, UAV trajectory, and AP's power distribution. Finally, we solved all subproblems by using convex optimization techniques, such as successive convex approximations (SCA), and proposed an iterative algorithm. In the simulation, we compared our proposed algorithm with different optimization schemes. The results indicated that our scheme outperforms the others. [ABSTRACT FROM AUTHOR]

This paper developed a new and efficient approach, the metaheuristic optimisation approach based on the Kalman filter (KF), to accurately identify the infinite impulse response (IIR) system coefficients. A recently developed metaheuristic optimisation technique called the manta-ray foraging optimisation (MRFO) algorithm has been employed to solve the proper parameters tuning exertion of KF. The problem of achieving better unknown state estimations by the conventional KF is closely connected with decent prior information about the noise statistics (measurement and process noise). However, the statistics mentioned above depend on the types of applications and process dynamics. The use of insufficiently known priori filter statistics would reduce the precision of the estimated filter or sometimes lead to a practical divergence of the filter. To alleviate this problem, the proposed technique initially produces optimally tuned KF parameters using the MRFO algorithm, and then the conventional KF method estimates the coefficients of the IIR model using the optimised KF parameters obtained at the beginning. The simulation studies have been carried out on three distinct order benchmark IIR systems (both same and reduced-order), and two prominent IIR applications called direct current (DC) motor and fractional order digital differentiator (FODD) using the proposed method. The overall identification results demonstrate that the proposed MRFO-KF technique significantly improves performance metrics over competing techniques, such as basic KF, other employed metaheuristic-based KF schemes and recently reported state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Particle swarm optimization approach (PSO) has illustrated promising performance on a variety of optimization problems. The critical issues that must be addressed in PSO research are how to balance conflicts around exploration, exploitation, and reasonably distribute computational resources. In this paper, a neighborhood-aware N states Markovian Switching PSO algorithm (NS-MJPSOloc) is proposed, where we have employed a local search based approach. The neighborhood of PSO algorithm is a single particle itself in the early stages of optimization. The neighborhood will eventually expand to include all particles as the amount of generations increases. The search space is segmented into N potential states according to the size of problem. The value of N is dependent on the dimension of problem space. Every iteration serves a transition probability-based Markovian switching technique to estimate the acceleration coefficients in an adaptable way. The transition probability determines the next state and adequate values for acceleration coefficients. Through adopting a local search strategy while interacting with all other particles in the group, every particle possesses a chance of improving itself. Moreover, new features have been incorporated to accomplish accuracy and stability. The primary contributions of the proposed work are to enhance the topological structure of a recently developed PSO variant by using a local search (loc) mechanism rather than a global (gbest) approach. The purpose of this idea is to develop a more adaptable and less sensitive novel version of PSO that can handle multi-modal and highly challenging problems. The proposed algorithm was then tested using a variety of uni-modal, multi-modal and rotated benchmark functions. The experimental findings show that the proposed NS-MJPSOloc outperformed most of the current versions in terms of accuracy and convergence time, notably for multi-modal problems. [ABSTRACT FROM AUTHOR]

The application of three different categories such as swarm-based, physics-based, and evolutionary-based techniques to various optimization problems related to different fields gained importance due to accuracy, speed and least chance to fall in local minima. Although there are number of physics based optimization techniques available to solve the optimization problems, however, they suffer from their complexity and may require multiple runs to achieve the optimum value. This paper focuses on a simple physics based problem which outperforms few of the existing algorithms. This work investigates the physics-based Lens formula integration using a population-based approach. The contender for a population-level solution to picture obstructions created by a convex, concave, or combination of the two lenses is the item with multiple dimensional locations. The object's location is updated using the image position. The fundamental goals of population-based optimization algorithms are highlighted by the concave, convex, or combination of both lenses, which enhances the speed of exploration, exploitation, and local optimum avoidance of the search space. The focal length equation, controls how the object's picture is produced, may be altered by specifying a number of random and adaptive factors to highlight the exploitation and exploration of the search space. The capacity of the proposed algorithm for exploitation, exploration, utilisation of the search space, avoidance of local minima, and attainment of global maxima is validated using unimodal, multimodal, and composite benchmark functions. In addition, a variety of real-world engineering issues taken into account to verify the performance of the proposed algorithm. The observations compared with few fundamental optimization techniques of the same category and found the proposed solution better performing in some context than the existing ones. [ABSTRACT FROM AUTHOR]

The transportation problems, which consist of multiple objectives with heterogeneous conveyances, are the pragmatic representation of the transportation occurring in the real world. However, situations do exist where the solutions obtained by the classical optimization techniques do not reflect the acumen of the decision-maker. Against this backdrop, a systematic algorithm is proposed in this paper to generate an initial population to tackle the multi-objective solid transportation problem efficiently. In addition, the non-dominated sorting genetic algorithm (NSGA) III is carried out to acquire solutions that can demonstrate the varying degrees of objectives. Furthermore, two problems of different sizes are framed, and their performance is compared via NSGA II, hybrid genetic algorithm, and fuzzy programming technique. [ABSTRACT FROM AUTHOR]

Cloud services with the property of elasticity are always bearing different loads over time. The efficiency of cloud management can be improved through the accurate prediction for cloud loads, which can be served as the foundation and recommendation for the future developing of the network scheme. As the basis and premise for management and decision making, a good prediction depends on suitable models and research methods. However, the interaction of users and the cloud network is high variability in the time and spatial dimensions, so traditional linear forecasting models cannot always process cloud computing loads well because of nonlinear changes. In this paper, a novel cloud load prediction model is proposed by improving whale optimization algorithm based on the hybrid strategy (HWOA) and combined with extreme learning machine (ELM) for strong nonlinear mapping ability. The proposed cloud load forecasting model is to employ HWOA optimizer to optimize the ELM model random parameters. Main research work includes that (1) the HWOA optimizer is to solve the whale optimizer local extremum problem; (2) the proposed HWOA optimizer reduces the ELM random parameters on cloud load forecasting; (3) the convergence performance is verified by benchmark testing functions; and (4) three groups of simulation experiments are conducted to evaluate the cloud load forecasting results. The results prove that the HWOA optimizer has a good convergence and outperforms the several conventional popular swarm intelligence optimizers. Also, the prediction results verify that our proposed model is more competitive for providing a solid foundation for efficient resource management and maximized economic benefits in cloud environment. For cloud load research groups and practitioners, this paper provides a new idea of cloud load forecasting. [ABSTRACT FROM AUTHOR]

In PV power generation systems, flexible power point tracking control needs to be introduced to regulate the output power on the PV side for meeting the grid connection requirements. However, traditional FPPT control methods tend to cause certain oscillations in the system. This paper proposes an FPPT method based on the adaptive lion swarm optimization algorithm (ALSO). First, an ALSO algorithm is proposed, which adds adaptive scaling parameters to the lion swarm optimization algorithm, so that the number of adult lions can be adjusted adaptively with increasing iterations to coordinate the local exploitation and global search abilities of the algorithm. Then the ALSO is used to selectively optimize the operation of different intervals so as to achieve the FPPT control on the left or right side of the maximum power point. The test experiments for the ALSO algorithm show that the ALSO algorithm has strong local exploitation and global search abilities, which is better than the compared algorithms. Through the simulations on the PV grid-connected system, the validity and superiority of the ALSO-based method over the traditional FPPT algorithm in terms of dynamic and steady-state performance are verified under different operating conditions. [ABSTRACT FROM AUTHOR]

Because of the complexity of data sets from the real world, it is difficult to classify the data sets clearly and effectively, thus we prefer to adopt fuzzy clustering approaches to analyze the data sets. However, due to the variety of fuzzy clustering algorithms, the different number of clusters will lead to different clustering results. The number of clusters is closely related to the clustering division, so how to determine the number of fuzzy clustering (k) has become a problem. Until now, many researchers have proposed utilizing fuzzy clustering validity indexes to deal with this kind of problem. However, the effectiveness index of fuzzy clustering can only be evaluated on the basis of the fuzzy clustering algorithm FCM to divide the clusters. When the range of k value is too large, FCM's clustering for different k values is quite time-consuming. From this perspective, this paper proposes a fuzzy clustering optimal k selection method based on multi-objective optimization (FMOEA-K). Different from the traditional methods, this method combines the fuzzy clustering effectiveness index with multi-objective optimization algorithm (MOEA), and uses multi-objective optimization algorithm to search the appropriate cluster center concurrently. Because of the concurrency of the multi-objective optimization algorithm, the calculation time is shortened. The experimental results show that compared with the traditional method, the FMOEA-K can shorten the calculation time and improve the accuracy of calculating the optimal k value. [ABSTRACT FROM AUTHOR]

The main purpose of this paper is to identify and segment each white blood cells (WBCs) from microscopic images and then classify it to affected or non-affected by acute lymphoblastic leukemia (ALL). The proposed model started by, firstly, detection and isolation of (WBCs). This was performed by conversion of cell images from RGB to CMYK color space. Then histogram equalization followed by thresholding estimation by Zack algorithm was performed for segmentation of cells from the surrounding blood contents. Secondly, some features were extracted from the segmented cells, and they included color, shape, texture and hybrid features. Thirdly, social spider optimization algorithm (SSOA) was applied to select the most appropriate features. Finally, several classifiers were used to validate the performance of the proposed algorithms. The proposed model was applied to the well-known ALL-IDB2 dataset. The results show: firstly, the segmentation (identification and isolations) results were 99.23, 100 and 97.1% of segmentation accuracy, sensitivity and specificity, respectively, which is the highest among other published papers. Secondly, the classification accuracy of the proposed model was higher than the non-SSOA in most experiments. The proposed system achieved remarkable results; 95.23% of classification accuracy with a feature reduction ratio of about 50%, which is the highest among the most recent works on the same dataset. [ABSTRACT FROM AUTHOR]

As a kind of particulate matter which is very detrimental to human health, the prediction of PM 2.5 has attracted increasing attention in air pollution forecasting and early warning. In this paper, density peaks clustering (DPC) is employed to select the individual models for combination forecasting, and a novel high-precision combination forecasting model named DPC-based combined model is proposed to predict PM 2.5 concentrations. Firstly, CEEMD is introduced to decompose the original P M 2.5 data into three intrinsic mode functions ( I M F s ) with different features. Then, 125 individual models are obtained by applying SVR and GRNN to these three I M F s , and the intelligent optimization algorithms are adopted to optimize the parameters of SVR. Finally, the 125 individual models are clustered by DPC based on five model evaluation indexes to select the individual model for combination forecasting. The experimental results of four Chinese cities demonstrate that the proposed model outperforms the optimal individual model and can greatly improve the prediction accuracy. Taking Dalian as an example, the MAE of the proposed model is 2.2 by clustering the 125 individual models into two groups, and the result reveals that the accuracy is increased by 65.73% compared with the minimum MAE (6.42) of the individual models. [ABSTRACT FROM AUTHOR]

The expanding approval of e-learning structure has made the need for the customized suggestion prototype which can be utilized to advance the successful learning condition for the learners. Customized suggestion model is a particular sort of data separating framework used to recognize a lot of articles that are applicable to a e-learners. In this paper, we mainly propose the efficient e-learning recommendation (EELR) system for user preferences using hybrid optimization algorithm (HOA). EELR system constructs a HOA with deep recurrent neural network (DRNN) and improved whale optimization (IWO) algorithm. First, DRNN is utilized to order the e-learner types dependent on these e-learner gatherings, clients can acquire course proposal from the gathering's persuasion. Thereafter, the conduct and the inclinations of the learners are examined via completing the mining of the arrangements watched every now and again by the IWO calculation. Rather than a learner effectively looking for data, recommender frameworks give counsel to students about articles they may wish to analyze. At last, the proposal of the e-learning depends on the appraisals comparing to these arrangements watched often. This proposed system is going to implement and validate in numerous e-learning entries against the client inclinations over some undefined time frame and demonstrated to be more proficiency and exactness contrasted with the customary recommender framework. This strategy can help learners to grasp the knowledge system and learning direction, and improve their learning efficiency. Observation results show that the proposed methodology empowers the asset suggestion to singular clients, which is started from different sources. [ABSTRACT FROM AUTHOR]

Multiple unmanned aerial vehicle (UAV) path planning is the benchmark problem of multiple UAV application, which belongs to the non-deterministic polynomial problem. Its objective is to require multiple UAV to fly safely to the goal position according to their specific start position in three-dimensional space. This issue can be defined as a high-dimensional optimization problem, the solution of which requires optimization techniques with global optimization capabilities. Equilibrium optimizer (EO) is a population-based meta-heuristic algorithm. To improve the optimization ability of EO to solve high-dimensional problems, this paper proposes a modified equilibrium optimizer with generalized opposition-based learning (MGOEO), which improves the population activity by increasing the internal mutation and cross of the population. In addition, the generalized opposition-based learning is used to construct the population, which can effectively ensure that the algorithm has ability to jump out of the limitation of local optimal. First, numerical experiments show that MGOEO has better optimization precision than EO and several other swarm intelligent algorithms. Then, the paths of UAVs are simulated in three different obstacle environments. The simulation results show that MGOEO can obtain safe and smooth paths, which are better than EO and other eight state-of-the-art optimization algorithms. [ABSTRACT FROM AUTHOR]

Teaching–learning-based optimization (TLBO) algorithm is an excellent human-inspired optimization technique. This paper proposes an innovative improved version of TLBO—collectivism teaching–learning-based optimization (CTLBO) algorithm. This algorithm imitates group and individual behaviours in the reality of teaching and learning, applies group-individual multi-mode cooperation strategies to form new search mechanism. The CTLBO contains three phases, i.e. preparation phase, teaching and learning phases. In the preparation phase, there are two operators, i.e. teacher self-learning and teacher-learner interaction operators. In the teaching phase, class teaching and performance-based group teaching operators are implied. In the learning phase, neighbour learning, student self-learning and team-learning strategies are mixed together to form three operators. Two sets of experiments are conducted to test the performance of CTLBO. The first set of experiments validates the improvement effect of CTLBO by comparing it with the original TLBO and other authors' improved versions of TLBO. The second set of experiments illustrates the advantage of CTLBO by comparing it with other 17 meta-heuristic algorithms in solving 30 general benchmark functions and 15 CEC2015 test suit functions. The results of experiments show that CTLBO algorithm has significant improvement effect compared with TLBO. It is the most effective one amongst the improved versions of TLBO selected for comparison, and outperforms all other 17 meta-heuristic algorithms. The algorithm can significantly improve the convergence ability and the accuracy in solving different-scale complex optimization models. [ABSTRACT FROM AUTHOR]

This paper presents a novel medical image processing technique for analyzing different peripheral blood cells such as monocytes, lymphocytes, neutrophils, eosinophils, basophils, and macrophages. However, the existing systems suffered from low accuracy while classifying the different blood cell images and also consume higher processing power. The proposed model consists of two major steps such as segmentation and classification of peripheral blood cells. The modified local-information weighted intuitionistic Fuzzy C-means clustering (MLWIFCM)-based golden eagle optimization algorithm performs the nucleus segmentation. Finally, the peripheral blood cell classes such as Basophil, Lymphocyte, Neutrophil, Monocyte, and Eosinophil are effectively classified using hybrid-parameter RNN-based remora optimization algorithm. The MATLAB R2019b is used as the implementation platform. To analyze the performances of our proposed method, we have taken two datasets; they are BCCD and LISC datasets. Meanwhile, the classification performances were analyzed with the aid of different performance metrics such as mean accuracy, mean intersection over union, mean average precision, and mean BF score values. [ABSTRACT FROM AUTHOR]

Segmentation refers to the process of dividing an image into multiple regions based on some criteria such as intensity and color. In recent years, color image segmentation has received considerable attention from the researchers. However, it is still a highly complicated task due to the presence of more attributes or components as compared to monochrome images. Numerous meta-heuristics algorithms are developed to determine the optimal threshold value for segmenting color images efficiently. This paper presents an enhanced sine cosine algorithm (ESCA) to seek threshold for segmenting color images. Sine cosine algorithm (SCA) is a population-based optimization algorithm which has the ability of preventing local minima problem. First an input image is transformed to CIE L*a*b* color reduced space. ESCA is applied to determine the optimal threshold values for segmentation. The performance of the proposed method is tested on color images from Berkeley database, and segmentation results are compared with two metaheuristic algorithms, namely particle swarm optimization (PSO) and standard SCA. Experimental results are validated by measuring peak signal–noise ratio (PSNR), structural similarity index and computation time for all the images investigated. Results revealed that the proposed method outperforms the other methods like PSO and SCA by achieving PSNR of 23 dB and SSIM of 0.93 and also require less time for finding optimal threshold values than PSO and SCA. [ABSTRACT FROM AUTHOR]

Cloud computing provides a wide variety of services, from small to big businesses, to individual consumers. Cloud computing's features entice users to migrate their operations from traditional platforms to cloud platforms. In comparison to traditional systems, cloud computing has an extremely powerful processing capacity. Requests for resources are considered tasks in the cloud, and appropriate resources are allocated depending on user needs. However, owing to high demand and volume of requests, cloud struggles to allocate resources. Task schedulers are employed in cloud computing to address these issues. Various task scheduling methods have been presented in several research publications, and the quest for a better scheduling model continues. In this paper, a task scheduling method based on a hybrid optimization algorithm is presented, which effectively schedules jobs with the least amount of waiting time. In addition to these, other parameters, such as the overall production time, execution time, waiting time, efficiency, and utilization are included in this study. The simulation results show that the proposed scheduling method is superior to conventional Ant Colony and Particle Swarm Optimization-based scheduling algorithms in terms of performance. [ABSTRACT FROM AUTHOR]

Traditional folk crafts are handicrafts and corresponding activities created by Chinese working people in all ages to meet their material and spiritual needs by using various material materials and handicrafts. It reflects the general characteristics of arts and crafts, but also shows the distinctive personality of national culture. The folk printing and dyeing process of Miao Nationality in Xiangxi and Southeast Guizhou has this characteristic, which once brought a lot of convenience to people's life and became an important part of people's life at that time. Traditional art carries a lot of spiritual things in its material form, so it has high esthetic value, cultural value and profound spiritual connotation. Therefore, the optimization design of traditional printing and dyeing process framework in Xiangxi is worth studying. In this paper, through using big data pattern mining algorithm and scene understanding algorithm, a framework optimization system of traditional printing and dyeing process in Xiangxi is constructed, for the optimization of the traditional model, the edge-driven scene model is applied for the systematic study. The designed model is simulated through the testing on different data sets. Reflecting from the experimental result, we can conclude that the proposed model is efficient. [ABSTRACT FROM AUTHOR]

Combined heat and power economic dispatch (CHPED) problem is one of the challenging issues in power system control, and operation, that searches for the optimal values of power and heat generation of different units to minimize the total operation cost. The interdependency of heat and power outputs of CHP units and valve point loading effects of power-only units make this problem very nonlinear and non-convex. So far, a wide variety of optimization algorithms based on simulation of various natural phenomena have been proposed and applied to solve this complex problem, claiming that the proposed algorithm is better than the similar optimization algorithms in terms of accuracy and run-time. In this paper, the CHPED problem is investigated in different case studies by using the imperialist competitive Harris hawks optimization (ICHHO) algorithm, as a combination of the imperialist competitive algorithm and Harris hawks optimization. The effectiveness of the proposed algorithm is first evaluated by testing it on the standard cases of 5, and 7 units as small-scale systems and 48-unit as the large-scale test system, considering the electrical losses. Also for the first time, the multi-zone CHPED (MZCHPED) problem is solved by a metaheuristic algorithm. The obtained results show that ICHHO algorithm reduces the operating costs in the range of 0.02–1.1%, compared to the best-reported results, in different single-zone CHPED (SZCHPED) problems. Also, the MZCHPED results, prove the effective performance of the algorithm compared to the similar SZCHPED problem. [ABSTRACT FROM AUTHOR]

The feed forward neural network (FNN) has a significant breakthrough in recent times in solving different real-life complex problems. The success of FNN is primarily attributed to its architecture, the optimization technique used, and the tuning of hyper-parameters to identify different patterns in data. This study tries to find an optimization technique that will play a vital role in an FNN to get better results. Whale optimization algorithm (WOA) is one of the popular nature-inspired optimization techniques used to solve real-time sciences and engineering problems. A new optimization approach namely chaotic oppositional-based whale optimization algorithm (COWOA) is developed in this paper, for the first time, to discuss the FNN problems of four different dataset by integrating chaotic function and oppositional-based learning with WOA. Simulation results are presented to validate the success of the suggested COWOA approach and demonstrate its dominance over basic WOA, chaotic WOA (CWOA), particle swarm optimization (PSO), Adam optimization algorithm (AOA), and stochastic gradient descent (SGD) for both unimodal and multimodal dataset problems. The comparative analysis with other optimization techniques validates its superiority and robustness on four standard dataset of FNN problem. Furthermore, to validate the performance of the proposed algorithm, a statistical analysis is performed. The outcome of the statistical results validates the superiority and acceptability of COWOA over other optimization methods. [ABSTRACT FROM AUTHOR]

A biomedical application of a novel metaheuristic optimizer is proposed in this paper by constructing an enhanced arithmetic optimization algorithm (AOA). The latter algorithm was constructed using the logarithmic spiral (Ls) search mechanism from the whale optimization algorithm and the greedy selection scheme from the differential evolution algorithm. The proposed algorithm (Ls-AOA) was tested against unimodal and multimodal benchmark functions and demonstrated better capability comparatively using other efficient metaheuristic algorithms reported in the literature. The constructed Ls-AOA algorithm was then proposed to design a proportional-integral-derivative (PID) controller employed in a functional electrical stimulation (FES) system for the first time. The initial statistical and convergence profile assessment showed better performance of the proposed algorithm. The comparative analyses for transient and frequency responses were performed for the PID-controlled FES system using the original AOA, sine–cosine and particle swarm optimization algorithms and the traditional Ziegler-Nichols tuning scheme. Similarly, the FES system tuned with the latter methods was also assessed for disturbance rejection and noise elimination. All the comparative analyses demonstrated that the proposed Ls-AOA has the greater capability for the challenging biomedical FES system. [ABSTRACT FROM AUTHOR]

In industrial applications and automation, the robotic manipulators exhibit a significant role. Several complex robotic systems performed a number of industrial works named spray painting, welding, assembly, pick and place action, etc. The end-effector's position and the joint angles play a vital role since any task is activated inside the pre-defined robotic manipulator's work space. Also, the problem of trajectory planning is a very challenging task in the robotic fields. To solve these problems, this paper proposes a kinematics analysis and trajectory planning of an Industrial robotic manipulators (IRMs) based on the hybrid optimization algorithms. Here, three IRMs such as PUMA 560 (6 DOF), KUKA LBR iiwa 14 R820 (7DOF) and ABB IRB 140 (6DOF) are considered. For each robot, the forward and the inverse kinematics (IK) are analysed and also the trajectory planning of each robot is discussed using the hybrid optimization algorithms. In this work, 18 optimization algorithms such as PSO (particle swarm optimization), SSO (social spider optimization), DFO (dragonfly optimization), BOA (butterfly optimization), CSA (crow search algorithm), BSA (bird swarm algorithm), SHO (selfish herd optimization), KHO (krill herd optimization), ALO (antlion optimization), ACO (ant colony optimization), GWO (Grey wolf optimization), GOA (grasshopper optimization), SBO (satin bowerbird optimizer), WCO (world cup optimization), COA (cuckoo optimization algorithm), CFA (cuttlefish algorithm), SOA (seagull optimization algorithm), and TSA (tunicate swarm algorithm) are utilized for both forward and inverse kinematic analysis and the trajectory planning problem. For the PUMA 560, KUKA LBR iiwa 14 R820, and ABB IRB 140 IRMs, forward kinematics (FK) are solved by the hybrid combination of PSO-SSO, SHO-KHO, and SBO-WCO individually. Also, the IK are solved by the DFO-BOA, ALO-ACO and COA-CFA for each IRMs. The trajectory planning problem is solved by the CSA-BSA, GWO-GOA and SOA-TSA for each robot individually. These optimization techniques give number of solution for kinematics and trajectory problem but it converses the best solution for the minimum multi-objective function value. Each robot obtained the minimum travelling time for without and with an obstacle which is 0.0118 and 0.0313 s for PUMA and 0.0117 and 0.0310 s for KUKA, and 0.0114 and 0.0120 s for ABB IRB 140 robot. The advantages of these hybrid algorithm are shorter computation time, and fewer iterations. The kinematics and trajectory analysis of each IRM is simulated using robotic tool box in MATLAB with GUI interface. For each IRM, optimized position values of the end effector, joint angles and the best optimal path are computed with minimum objective function. Finally, the performance of each robot is compared to the existing robotic works. [ABSTRACT FROM AUTHOR]

In this paper, the combination of particle swarm optimization (PSO) and gravitational search algorithm (GSA) is enhanced by the first-order gradient method and a new optimization algorithm is introduced as GPSG. In metaheuristic methods, some search directions are randomly selected and the resulting information gradually progresses toward the optimal solution. Since along the gradient direction usually yields the largest decrease in the desired function, it is added to the GSA and PSO process to allow for faster and more accurate convergence. By integrating the metaheuristic methods with the gradient directions, a powerful method for optimizing functions has been made possible. A novel approach is introduced by merging two metaheuristic methods with the randomly generated steepest decent directions. Numerous examples of unconstrained problems of mathematical functions of CEC2005 and CEC2017 together with some constrained examples of stress and displacement structural design problems have been chosen to demonstrate the reliability and capability of the presented method. Comparison of the numerical results with some methods indicates that the average rank of the proposed technique is better. [ABSTRACT FROM AUTHOR]

The field of Metaheuristics has produced a large number of algorithms for continuous, black-box optimization. In contrast, there are few standard benchmark problem sets, limiting our ability to gain insight into the empirical performance of these algorithms. Clustering problems have been used many times in the literature to evaluate optimization algorithms. However, much of this work has occurred independently on different problem instances and the various experimental methodologies used have produced results which are frequently incomparable and provide little knowledge regarding the difficulty of the problems used, or any platform for comparing and evaluating the performance of algorithms. This paper discusses sum of squares clustering problems from the optimization viewpoint. Properties of the fitness landscape are analysed and it is proposed that these problems are highly suitable for algorithm benchmarking. A set of 27 problem instances (from 4-D to 40-D), based on three well-known datasets, is specified. Baseline experimental results are presented for the Covariance Matrix Adaptation-Evolution Strategy and several other standard algorithms. A web-repository has also been created for this problem set to facilitate future use for algorithm evaluation and comparison. [ABSTRACT FROM AUTHOR]

The solution of ordinary differential equations (ODEs) arises in a wide variety of engineering problems. This paper presents a novel method for the numerical solution of ODEs using improved artificial neural networks (IANNs). In the first step, we derive an approximate solution of ODEs by artificial neural networks (ANNs). Then, we construct a joint cost function of network system, it consists of several error functions corresponding to different sample points, and we reformulate Levenberg–Marquardt (RLM) algorithm to adjust the network parameters. The advantages of this method are high calculation accuracy and fast convergence speed compared with other existed methods, also increasing the simulation stability of ANNs method. The performance of the new proposed method in terms of calculation accuracy and convergence speed is analyzed for several different types of nonlinear ODEs. [ABSTRACT FROM AUTHOR]

With the increasing hazard to the environment and progressions in renewable energy technologies, hybrid renewable energy systems can meet the energy demand. Total dependability can be attained by adding battery banks to hybrid systems. This paper aims to optimally design a multi-source grid isolated hybrid generation system for future smart cities in the state of Tamil Nadu, India. This off-grid power generation consists of wind turbine generators, photovoltaic panels, inverters, diesel generators, and batteries. The design intents at minimizing Net Present Cost, Unmet Load, and CO2 emissions which are conventionally contradictory to each other. This paper analyzes the optimal off-grid combination of components and control strategies, utilizing improved Hybrid Optimization using the Genetic Algorithm. The results obtained portrays that a mix of hybrid renewable energy generators at off-grid locations without diesel generators can be a cost-effective choice of new smart cities, and it is sustainable, environmentally viable, and techno-economically feasible. Subvention is imperative to take on the large-scale system due to the high capital cost. [ABSTRACT FROM AUTHOR]

This paper proposes a novel control-centric optimization algorithm, i.e., proportional–integral–derivative optimization algorithm. The proposed optimization algorithm is inspired by the conventional proportional–integral–derivative controller. The proposed optimization algorithm consists of two types of controllers, i.e., explorative controllers with variable parameters and exploitative controllers with fixed parameters. In the exploration process of the approach, multiple explorative controllers with variable parameters move toward the global optimal solution domain. In the exploitation process of the approach, multiple exploitative controllers with fixed parameters moving toward to local optimal solution. The case studies results obtained by the proposed proportional–integral–derivative optimization algorithm under a total of eight basic mathematical optimization problems and the parameters optimization problem of the double-fed induction generator with the maximum wind power tracking technique show that the proportional–integral–derivative optimization algorithm can explore and exploit the global optimal solution effectively. [ABSTRACT FROM AUTHOR]

Training optimization plays a vital role in the development of convolution neural network (CNN). CNNs are hard to train because of the presence of multiple local minima. The optimization problem for a CNN is non-convex, hence, has multiple local minima. If any of the chosen hyper-parameters are not appropriate, it will end up at bad local minima, which leads to poor performance. Hence, proper optimization of the training algorithm for CNN is the key to converge to a good local minimum. Therefore, in this paper, we introduce an evolutionary convolution neural network (ModPSO-CNN) algorithm. The proposed algorithm results in the fusion of modified particle swarm optimization (ModPSO) along with backpropagation (BP) and convolution neural network (CNN). The training of CNN involves ModPSO along with backpropagation (BP) algorithm to encourage performance improvement by avoiding premature convergence and local minima. The ModPSO have adaptive, dynamic and improved parameters, to handle the issues in training CNN. The adaptive and dynamic parameters bring a proper balance between the global and local search ability, while an improved parameter keeps the diversity of the swarm. The proposed ModPSO algorithm is validated on three standard mathematical test functions and compared with three variants of the benchmark PSO algorithm. Furthermore, the performance of the proposed ModPSO-CNN is also compared with other training algorithms focusing on the analysis of computational cost, convergence and accuracy based on a standard problem specific to classification applications, such as CIFAR-10 dataset and face and skin detection dataset. [ABSTRACT FROM AUTHOR]

An efficient optimization of network weights has been the primary goal of the artificial neural network (ANN) research community since decades. The aim of every optimization problem is to minimize the network cost which is some form of error function between the desired and the actual network outputs, during the training phase. The conventional gradient-based optimization algorithms like backpropagation are likely to get trapped in local minima and are sensitive to choices of initial weights. The evolutionary algorithms have proved their usefulness in introducing randomness into the optimization procedure, since they work on a global search strategy and induce a globally minimum solution for the network weights. In this paper, we particularly focus on ANN trained by Particle Swarm Optimization (ANN-PSO), in which the local-best and global-best particle positions represent possible solutions to the set of network weights. The global-best position of the swarm, which corresponds to the minimum cost function over time, is determined in our work by minimizing a new non-extensive cross-entropy error cost function. The non-extensive cross-entropy is derived from the non-extensive entropy with Gaussian gain that has proven to give minimum values for regular textures containing periodic information represented by uneven probability distributions. The new cross-entropy is defined, and its utility for optimizing the network weights to a globally minimum solution is analyzed in this paper. Extensive experimentation on two different versions: the baseline ANN-PSO and one of its most recent variants IOPSO-BPA, on benchmark datasets from the UCI repository, with comparisons to the state of the art, validates the efficacy of our method. [ABSTRACT FROM AUTHOR]

In this paper, the problem of safe global maximization (it should not be confused with robust optimization) of expensive noisy black-box functions satisfying the Lipschitz condition is considered. The notion "safe" means that the objective function f(x) during optimization should not violate a "safety" threshold, for instance, a certain a priori given value h in a maximization problem. Thus, any new function evaluation (possibly corrupted by noise) must be performed at "safe points" only, namely, at points y for which it is known that the objective function f (y) > h . The main difficulty here consists in the fact that the used optimization algorithm should ensure that the safety constraint will be satisfied at a point ybefore evaluation of f(y) will be executed. Thus, it is required both to determine the safe region Ω within the search domain D and to find the global maximum within Ω . An additional difficulty consists in the fact that these problems should be solved in the presence of the noise. This paper starts with a theoretical study of the problem, and it is shown that even though the objective function f(x) satisfies the Lipschitz condition, traditional Lipschitz minorants and majorants cannot be used due to the presence of the noise. Then, a δ -Lipschitz framework and two algorithms using it are proposed to solve the safe global maximization problem. The first method determines the safe area within the search domain, and the second one executes the global maximization over the found safe region. For both methods, a number of theoretical results related to their functioning and convergence is established. Finally, numerical experiments confirming the reliability of the proposed procedures are performed. [ABSTRACT FROM AUTHOR]

Forming a team of experts that can match the requirements of a collaborative task is an important aspect, especially in project development. In this paper, we propose an improved Jaya optimization algorithm for minimizing the communication cost among team experts to solve team formation problem. The proposed algorithm is called an improved Jaya algorithm with a modified swap operator (IJMSO). We invoke a single-point crossover in the Jaya algorithm to accelerate the search, and we apply a new swap operator within Jaya algorithm to verify the consistency of the capabilities and the required skills to carry out the task. We investigate the IJMSO algorithm by implementing it on two real-life datasets (i.e., digital bibliographic library project and StackExchange) to evaluate the accuracy and efficiency of proposed algorithm against other meta-heuristic algorithms such as genetic algorithm, particle swarm optimization, African buffalo optimization algorithm and standard Jaya algorithm. Experimental results suggest that the proposed algorithm achieves significant improvement in finding effective teams with minimum communication costs among team members for achieving the goal. [ABSTRACT FROM AUTHOR]

Based on uncertainty theory, this paper mainly studies the uncertainties and the information value in the two-stage uncertain programming with recourse. We first define three fundamental concepts and investigate their theoretical properties, based on which we present two optimal indices, i.e., EVPI and VUS. Then, we introduce a method to calculate the expected value of the second-stage objective function involving discrete uncertain variables. Due to the complexity of calculation, the upper bound and lower bound for the two indices are studied, respectively. Finally, two examples are given to illustrate these concepts clearly. The results obtained in this paper can provide theoretical basis for studying uncertainties and information value in decision-making process under uncertain systems. [ABSTRACT FROM AUTHOR]

Regression analysis is a method to estimate the relationships among the response variable and the explanatory variables. Assuming the observations of the response variable are imprecise and modeling the observed data via uncertain variables, this paper explores an approach of uncertain regression analysis to estimating the relationships among the variables with imprecisely observed samples. On the principle of least squares, an optimization problem is derived to calculate the unknown parameters in the regression model. In particular, this paper investigates uncertain linear regression model and gives an analytic representation of the unknown parameters. [ABSTRACT FROM AUTHOR]

A synergistic perspective to selecting the right project portfolio is the key to realizing strategic organizational objectives. Based on the theory of the composite system, this paper identifies the factors that affect the science of project portfolio selection (PPS) and investigates the synergistic relationship among the organizational strategic objective subsystem, the project portfolio subsystem and the environment subsystem, which collectively constitute the composite system of project portfolio selection. Utilizing the concept of order parameters, an order parameter system is built for each subsystem to reflect the synergistic relationships in the composite system of PPS. The PPS model is constructed based on the degree of synergy in the composite system by describing the order degree of the subsystem and the synergy degree of the composite system, which helps to quantify the synergistic relationships among these subsystems. The effectiveness and feasibility of the proposed model are demonstrated and validated by a case study. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the possibility of covariate shift adaptation in off-policy temporal difference learning for the class of fast mixing Markov chains. Off-policy evaluation algorithms in reinforcement learning such as off-policy least squares temporal difference (LSTD) deal with the problem of evaluating a target policy different from the sampling (or behavior) policy. Off-policy LSTD may result in poor quality of solution due to the shift among stationary distributions of the chains induced by following the target and behavior policies. Previous works—least squares temporal difference–distribution optimization (LSTD-DO) and the recently proposed emphatic TD—each tackles this problem by mapping distribution of states collected following the behavior policy (i.e. off-policy samples) to a new different distribution with better LSTD solution. In this paper, we consider off-policy LSTD in the class of target Markov chains with fast mixing time. For this class of problems, we propose adapting the distribution of off-policy state samples to the distribution of state samples after transition model adaptation, using a regularized covariate shift adaptation algorithm called least squares importance fitting. Empirical evaluations of our proposed approach on two classes of fast mixing chains show promising results in comparison with LSTD-DO and unadapted off-policy LSTD as the number of samples increases. [ABSTRACT FROM AUTHOR]

Discretization is an important preprocessing technique, especially in classification problems. It reduces and simplifies data, accelerates the learning process, and improves learner performance. The most challenging aspect of the discretization process is to maintain the accuracy of the classification algorithm and to prevent information loss while reducing the number of discretized values. In this paper, using evolutionary multi-objective optimization, classification error (the first objective function) and number of cut points (the second objective function) are simultaneously reduced. The third objective function involves selecting low-frequency cut points so that a smaller degree of information is lost during this conversion (from continuous to discrete). To the best of our knowledge, this is the first paper to consider the discretization process as a multi-objective optimization problem. Previous discretization methods result in only one solution. However, in real-world problems, decision makers often need several alternatives to make better decisions-a requirement which cannot be fulfilled using these techniques. The multi-objective nature of the proposed algorithm enables the generation of numerous solutions (i.e., the Pareto front) allowing the user to select the most appropriate solution according to the nuances of the problem. A total of 20 benchmark data sets were used to test the performance of the proposed algorithm. Our results show that the proposed algorithm offers superior performance compared to other methods in the literature. Thus, it presents better discretization in classification problems. [ABSTRACT FROM AUTHOR]