Deep learning attentionmechanisms have achieved remarkable progress in computer vision,but still face limitations when handling images with ambiguous boundaries and uncertain feature representations.Conventional atten...Deep learning attentionmechanisms have achieved remarkable progress in computer vision,but still face limitations when handling images with ambiguous boundaries and uncertain feature representations.Conventional attention modules such as SE-Net,CBAM,ECA-Net,and CA adopt a deterministic paradigm,assigning fixed scalar weights to features without modeling ambiguity or confidence.To overcome these limitations,this paper proposes the Fuzzy Attention Network Layer(FANL),which integrates intuitionistic fuzzy set theory with convolutional neural networks to explicitly represent feature uncertainty through membership(μ),non-membership(ν),and hesitation(π)degrees.FANLconsists of four coremodules:(1)feature dimensionality reduction via global pooling,(2)fuzzymodeling using learnable clustering centers,(3)adaptive attention generation through weighted fusion of fuzzy components,and(4)feature refinement through residual connections.A cross-layer guidance mechanism is further introduced to enhance hierarchical feature propagation,allowing high-level semantic features to incorporate fine-grained texture information from shallow layers.Comprehensive experiments on three benchmark datasets—PathMNIST-30000,full PathMNIST,and Blood MNIST—demonstrate the effectiveness and generalizability of FANL.The model achieves 84.41±0.56%accuracy and a 1.69%improvement over the baseline CNN while maintaining lightweight computational complexity.Ablation studies show that removing any component causes a 1.7%–2.0%performance drop,validating the synergistic contribution of each module.Furthermore,FANL provides superior uncertainty calibration(ECE=0.0452)and interpretable selective prediction under uncertainty.Overall,FANL presents an efficient and uncertaintyaware attention framework that improves both accuracy and reliability,offering a promising direction for robust visual recognition under ambiguous or noisy conditions.展开更多
The sustainability of the Internet of Things(IoT)involves various issues,such as poor connectivity,scalability problems,interoperability issues,and energy inefficiency.Although the Sixth Generation of mobile networks(...The sustainability of the Internet of Things(IoT)involves various issues,such as poor connectivity,scalability problems,interoperability issues,and energy inefficiency.Although the Sixth Generation of mobile networks(6G)allows for Ultra-Reliable Low-Latency Communication(URLLC),enhanced Mobile Broadband(eMBB),and massive Machine-Type Communications(mMTC)services,it faces deployment challenges such as the short range of sub-THz and THz frequency bands,low capability to penetrate obstacles,and very high path loss.This paper presents a network architecture to enhance the connectivity of wireless IoT mesh networks that employ both 6G and Wi-Fi technologies.In this architecture,local communications are carried through the mesh network,which uses a virtual backbone to relay packets to local nodes,while remote communications are carried through the 6G network.The virtual backbone is created using a heuristic distributed ConnectedDominating Set(CDS)algorithm.In this algorithm,each node uses information collected from its one-and two-hop neighbors to determine its role and find the set of expansion nodes that are used to select the next CDS nodes.The proposed algorithm has O(n)message and O(K)time complexities,where n is the number of nodes in the network,and K is the depth of the cluster.The study proved that the approximation ratio of the algorithmhas an upper bound of 2.06748(3.4306MCDS+4.8185).Performance evaluations compared the size of the CDS against the theoretical limit and recent CDS clustering algorithms.Results indicate that the proposed algorithm has the smallest average slope for the size of the CDS as the number of nodes increases.展开更多
Circumlunar abort trajectories constitute a vital contingency return strategy during the translunar phase of crewed lunar missions.This paper proposes a methodology for constructing the solution set of the circumlunar...Circumlunar abort trajectories constitute a vital contingency return strategy during the translunar phase of crewed lunar missions.This paper proposes a methodology for constructing the solution set of the circumlunar abort trajectory and leverages its advantageous properties to address the optimization design problem of abort trajectories.Initially,a solution set of all feasible abort trajectories,originating from an abort point on the nominal trajectory and complying with fundamental reentry constraints,is formulated through the introduction of two novel design parameters.Subsequently,the geometric characteristics of the solution set,as well as the distributional properties of key iterative constraint responses,including flight time and velocity increment,are analyzed.Finally,the characteristics exhibited in the solution set are employed to directly identify the design parameters of the abort trajectories with minimum flight time and velocity increment,thereby providing solutions to two distinct types of optimization problems.The simulation results for a variety of nominal trajectories,encompassing the reconstruction and redesign of the Apollo13 abort trajectory,validate the proposed method,demonstrating its ability to directly generate optimal abort trajectories.The method proposed in this paper investigates feasible abort trajectories from a global perspective,providing both a framework and convenience for mission planning and iterative optimization in abort trajectory design.展开更多
A discrete subset S of a topological gyrogroup G with the identity 0 is said to be a suitable set for G if it generates a dense subgyrogroup of G and S∪{0}is closed in G.In this paper,it is proved that each countable...A discrete subset S of a topological gyrogroup G with the identity 0 is said to be a suitable set for G if it generates a dense subgyrogroup of G and S∪{0}is closed in G.In this paper,it is proved that each countable Hausdorff topological gyrogroup has a suitable set;moreover,it is shown that each separable metrizable strongly topological gyrogroup has a suitable set.展开更多
Objectives:Tamoxifen is a key drug that provides endocrine therapy for estrogen receptor(ER)α-positive breast cancer;however,resistance remains a significant clinical challenge.This study aims to investigate the mole...Objectives:Tamoxifen is a key drug that provides endocrine therapy for estrogen receptor(ER)α-positive breast cancer;however,resistance remains a significant clinical challenge.This study aims to investigate the molecular mechanisms of tamoxifen resistance in ERα-positive breast cancer,with particular focus on the role of SET Domain Containing 1A(SETD1A)-driven forkhead box A2(FOXA2)as a key regulator of this resistance.Methods:FOXA2 expression and its regulation by SETD1A were assessed via(quantitative polymerase chain reaction),western blotting,transcriptome profiling,and chromatin immunoprecipitation analyses.The effects of FOXA2 on cell proliferation,migration,invasion,and cancer stem cell traits were evaluated using small interfering RNA(siRNA)-mediated silencing.Clinical relevance was examined by analyzing patient datasets and tumor tissue microarrays.Results:FOXA2 expression was significantly elevated in tamoxifen-resistant(TamR)and ERα-negative breast cancer cells compared to that in ERα-positive MCF-7 cells,regardless of tamoxifen treatment or ERαdepletion.Transcriptome and chromatin immunoprecipitation analyses revealed that SETD1A,a histone methyltransferase,directly regulated FOXA2 expression.Functionally,FOXA2 knockdown inhibited the proliferation,migration,invasion,and cancer stem cell properties of TamR cells while restoring tamoxifen sensitivity.High FOXA2 expression was correlated with poor survival and reduced responsiveness to tamoxifen in patients with ER-positive breast cancer.Conclusion:Our findings identified FOXA2 as a key mediator of tamoxifen resistance regulated by SETD1A and suggested that targeting the SETD1A-FOXA2 axis may offer a novel strategy for overcoming endocrine resistance in breast cancer.展开更多
The quality of cardiopulmonary resuscitation(CPR) significantly influences survival and neurological outcomes in patients with cardiac arrest(CA).Although mechanical chest compression devices and extracorporeal cardio...The quality of cardiopulmonary resuscitation(CPR) significantly influences survival and neurological outcomes in patients with cardiac arrest(CA).Although mechanical chest compression devices and extracorporeal cardiopulmonary resuscitation(ECPR) have demonstrated some benefits,high-quality manual CPR remained the essential first step,particularly in resource-limited settings.In this study,we examined whether opportunities existed to improve manual CPR performance using preliminary data from our recent survey conducted in a province in western China.We aim to emphasize the importance of improving manual CPR quality before implementing advanced interventions.展开更多
The non-dominated sorting genetic algorithm （NSGA） is improved with the controlled elitism and dynamic crowding distance. A novel multi-objective optimization algorithm is obtained for wind turbine blades. As an exa...The non-dominated sorting genetic algorithm （NSGA） is improved with the controlled elitism and dynamic crowding distance. A novel multi-objective optimization algorithm is obtained for wind turbine blades. As an example, a 5 MW wind turbine blade design is presented by taking the maximum power coefficient and the minimum blade mass as the optimization objectives. The optimal results show that this algorithm has good performance in handling the multi-objective optimization of wind turbines, and it gives a Pareto-optimal solution set rather than the optimum solutions to the conventional multi objective optimization problems. The wind turbine blade optimization method presented in this paper provides a new and general algorithm for the multi-objective optimization of wind turbines.展开更多
In order to solve discrete multi-objective optimization problems, a non-dominated sorting quantum particle swarm optimization (NSQPSO) based on non-dominated sorting and quantum particle swarm optimization is proposed...In order to solve discrete multi-objective optimization problems, a non-dominated sorting quantum particle swarm optimization (NSQPSO) based on non-dominated sorting and quantum particle swarm optimization is proposed, and the performance of the NSQPSO is evaluated through five classical benchmark functions. The quantum particle swarm optimization (QPSO) applies the quantum computing theory to particle swarm optimization, and thus has the advantages of both quantum computing theory and particle swarm optimization, so it has a faster convergence rate and a more accurate convergence value. Therefore, QPSO is used as the evolutionary method of the proposed NSQPSO. Also NSQPSO is used to solve cognitive radio spectrum allocation problem. The methods to complete spectrum allocation in previous literature only consider one objective, i.e. network utilization or fairness, but the proposed NSQPSO method, can consider both network utilization and fairness simultaneously through obtaining Pareto front solutions. Cognitive radio systems can select one solution from the Pareto front solutions according to the weight of network reward and fairness. If one weight is unit and the other is zero, then it becomes single objective optimization, so the proposed NSQPSO method has a much wider application range. The experimental research results show that the NSQPS can obtain the same non-dominated solutions as exhaustive search but takes much less time in small dimensions; while in large dimensions, where the problem cannot be solved by exhaustive search, the NSQPSO can still solve the problem, which proves the effectiveness of NSQPSO.展开更多
Through the transformation of hydraulic constraints into the objective functions associated with a water supply network rehabilitation problem, a non-dominated sorting Genetic Algorithm-II (NSGA-II) can be used to sol...Through the transformation of hydraulic constraints into the objective functions associated with a water supply network rehabilitation problem, a non-dominated sorting Genetic Algorithm-II (NSGA-II) can be used to solve the altered multi-objective optimization model. The introduction of NSGA-II into water supply network optimal rehabilitation problem solves the conflict between one fitness value of standard genetic algorithm (SGA) and multi-objectives of rehabilitation problem. And the uncertainties brought by using weight coefficients or punish functions in conventional methods are controlled. And also by in-troduction of artificial inducement mutation (AIM) operation, the convergence speed of population is accelerated;this operation not only improves the convergence speed, but also improves the rationality and feasibility of solutions.展开更多
Vehicle routing problem in distribution(VRPD)is a widely used type of vehicle routing problem(VRP),which has been proved as NP-Hard,and it is usually modeled as single objective optimization problem when modeling.For ...Vehicle routing problem in distribution(VRPD)is a widely used type of vehicle routing problem(VRP),which has been proved as NP-Hard,and it is usually modeled as single objective optimization problem when modeling.For multi-objective optimization model,most researches consider two objectives.A multi-objective mathematical model for VRP is proposed,which considers the number of vehicles used,the length of route and the time arrived at each client.Genetic algorithm is one of the most widely used algorithms to solve VRP.As a type of genetic algorithm(GA),non-dominated sorting in genetic algorithm-Ⅱ(NSGA-Ⅱ)also suffers from premature convergence and enclosure competition.In order to avoid these kinds of shortage,a greedy NSGA-Ⅱ(GNSGA-Ⅱ)is proposed for VRP problem.Greedy algorithm is implemented in generating the initial population,cross-over and mutation.All these procedures ensure that NSGA-Ⅱis prevented from premature convergence and refine the performance of NSGA-Ⅱat each step.In the distribution problem of a distribution center in Michigan,US,the GNSGA-Ⅱis compared with NSGA-Ⅱ.As a result,the GNSGA-Ⅱis the most efficient one and can get the most optimized solution to VRP problem.Also,in GNSGA-Ⅱ,premature convergence is better avoided and search efficiency has been improved sharply.展开更多
With the continuous increase of rapid urbanization and population growth,sustainable urban land-use planning is becoming a more complex and challenging task for urban planners and decision-makers.Multi-objective land-...With the continuous increase of rapid urbanization and population growth,sustainable urban land-use planning is becoming a more complex and challenging task for urban planners and decision-makers.Multi-objective land-use allocation can be regarded as a complex spatial optimization problem that aims to achieve the possible trade-offs among multiple and conflicting objectives.This paper proposes an improved Non-dominated Sorting Biogeography-Based Optimization(NSBBO)algorithm for solving the multi-objective land-use allocation problem,in which maximum accessibility,maximum compactness,and maximum spatial integration were formulated as spatial objectives;and space syntax analysis was used to analyze the potential movement patterns in the new urban planning area of the city of Kigali,Rwanda.Efficient Non-dominated Sorting(ENS)algorithm and crossover operator were integrated into classical NSBBO to improve the quality of non-dominated solutions,and local search ability,and to accelerate the convergence speed of the algorithm.The results showed that the proposed NSBBO exhibited good optimal solutions with a high hypervolume index compared to the classical NSBBO.Furthermore,the proposed algorithm could generate optimal land use scenarios according to the preferred objectives,thus having the potential to support the decision-making of urban planners and stockholders in revising and updating the existing detailed master plan of land use.展开更多
This paper uses the Butterfly Optimization Algorithm(BOA)with dominated sorting and crowding distance mechanisms to solve multi-objective optimization problems.There is also an improvement to the original version of B...This paper uses the Butterfly Optimization Algorithm(BOA)with dominated sorting and crowding distance mechanisms to solve multi-objective optimization problems.There is also an improvement to the original version of BOA to alleviate its drawbacks before extending it into a multi-objective version.Due to better coverage and a well-distributed Pareto front,non-dominant rankings are applied to the modified BOA using the crowding distance strategy.Seven benchmark functions and eight real-world problems have been used to test the performance of multi-objective non-dominated advanced BOA(MONSBOA),including unconstrained,constrained,and real-world design multiple-objective,highly nonlinear constraint problems.Various performance metrics,such as Generational Distance(GD),Inverted Generational Distance(IGD),Maximum Spread(MS),and Spacing(S),have been used for performance comparison.It is demonstrated that the new MONSBOA algorithm is better than the compared algorithms in more than 80%occasions in solving problems with a variety of linear,nonlinear,continuous,and discrete characteristics based on the Pareto front when compared quantitatively.From all the analysis,it may be concluded that the suggested MONSBOA is capable of producing high-quality Pareto fronts with very competitive results with rapid convergence.展开更多
In computer vision,convolutional neural networks have a wide range of uses.Images representmost of today’s data,so it’s important to know how to handle these large amounts of data efficiently.Convolutional neural ne...In computer vision,convolutional neural networks have a wide range of uses.Images representmost of today’s data,so it’s important to know how to handle these large amounts of data efficiently.Convolutional neural networks have been shown to solve image processing problems effectively.However,when designing the network structure for a particular problem,you need to adjust the hyperparameters for higher accuracy.This technique is time consuming and requires a lot of work and domain knowledge.Designing a convolutional neural network architecture is a classic NP-hard optimization challenge.On the other hand,different datasets require different combinations of models or hyperparameters,which can be time consuming and inconvenient.Various approaches have been proposed to overcome this problem,such as grid search limited to low-dimensional space and queuing by random selection.To address this issue,we propose an evolutionary algorithm-based approach that dynamically enhances the structure of Convolution Neural Networks(CNNs)using optimized hyperparameters.This study proposes a method using Non-dominated sorted genetic algorithms(NSGA)to improve the hyperparameters of the CNN model.In addition,different types and parameter ranges of existing genetic algorithms are used.Acomparative study was conducted with various state-of-the-art methodologies and algorithms.Experiments have shown that our proposed approach is superior to previous methods in terms of classification accuracy,and the results are published in modern computing literature.展开更多
文摘Deep learning attentionmechanisms have achieved remarkable progress in computer vision,but still face limitations when handling images with ambiguous boundaries and uncertain feature representations.Conventional attention modules such as SE-Net,CBAM,ECA-Net,and CA adopt a deterministic paradigm,assigning fixed scalar weights to features without modeling ambiguity or confidence.To overcome these limitations,this paper proposes the Fuzzy Attention Network Layer(FANL),which integrates intuitionistic fuzzy set theory with convolutional neural networks to explicitly represent feature uncertainty through membership(μ),non-membership(ν),and hesitation(π)degrees.FANLconsists of four coremodules:(1)feature dimensionality reduction via global pooling,(2)fuzzymodeling using learnable clustering centers,(3)adaptive attention generation through weighted fusion of fuzzy components,and(4)feature refinement through residual connections.A cross-layer guidance mechanism is further introduced to enhance hierarchical feature propagation,allowing high-level semantic features to incorporate fine-grained texture information from shallow layers.Comprehensive experiments on three benchmark datasets—PathMNIST-30000,full PathMNIST,and Blood MNIST—demonstrate the effectiveness and generalizability of FANL.The model achieves 84.41±0.56%accuracy and a 1.69%improvement over the baseline CNN while maintaining lightweight computational complexity.Ablation studies show that removing any component causes a 1.7%–2.0%performance drop,validating the synergistic contribution of each module.Furthermore,FANL provides superior uncertainty calibration(ECE=0.0452)and interpretable selective prediction under uncertainty.Overall,FANL presents an efficient and uncertaintyaware attention framework that improves both accuracy and reliability,offering a promising direction for robust visual recognition under ambiguous or noisy conditions.
基金Deputyship for Research&Innovation,Ministry of Education in Saudi Arabia for funding this research work through the project number RI-44-0028.
文摘The sustainability of the Internet of Things(IoT)involves various issues,such as poor connectivity,scalability problems,interoperability issues,and energy inefficiency.Although the Sixth Generation of mobile networks(6G)allows for Ultra-Reliable Low-Latency Communication(URLLC),enhanced Mobile Broadband(eMBB),and massive Machine-Type Communications(mMTC)services,it faces deployment challenges such as the short range of sub-THz and THz frequency bands,low capability to penetrate obstacles,and very high path loss.This paper presents a network architecture to enhance the connectivity of wireless IoT mesh networks that employ both 6G and Wi-Fi technologies.In this architecture,local communications are carried through the mesh network,which uses a virtual backbone to relay packets to local nodes,while remote communications are carried through the 6G network.The virtual backbone is created using a heuristic distributed ConnectedDominating Set(CDS)algorithm.In this algorithm,each node uses information collected from its one-and two-hop neighbors to determine its role and find the set of expansion nodes that are used to select the next CDS nodes.The proposed algorithm has O(n)message and O(K)time complexities,where n is the number of nodes in the network,and K is the depth of the cluster.The study proved that the approximation ratio of the algorithmhas an upper bound of 2.06748(3.4306MCDS+4.8185).Performance evaluations compared the size of the CDS against the theoretical limit and recent CDS clustering algorithms.Results indicate that the proposed algorithm has the smallest average slope for the size of the CDS as the number of nodes increases.
文摘Circumlunar abort trajectories constitute a vital contingency return strategy during the translunar phase of crewed lunar missions.This paper proposes a methodology for constructing the solution set of the circumlunar abort trajectory and leverages its advantageous properties to address the optimization design problem of abort trajectories.Initially,a solution set of all feasible abort trajectories,originating from an abort point on the nominal trajectory and complying with fundamental reentry constraints,is formulated through the introduction of two novel design parameters.Subsequently,the geometric characteristics of the solution set,as well as the distributional properties of key iterative constraint responses,including flight time and velocity increment,are analyzed.Finally,the characteristics exhibited in the solution set are employed to directly identify the design parameters of the abort trajectories with minimum flight time and velocity increment,thereby providing solutions to two distinct types of optimization problems.The simulation results for a variety of nominal trajectories,encompassing the reconstruction and redesign of the Apollo13 abort trajectory,validate the proposed method,demonstrating its ability to directly generate optimal abort trajectories.The method proposed in this paper investigates feasible abort trajectories from a global perspective,providing both a framework and convenience for mission planning and iterative optimization in abort trajectory design.
基金supported by Fujian Provincial Natural Science Foundation of China(2024J02022)the NSFC(11571158)+1 种基金supported by the NSFC(12071199)supported by the Young and middle-aged project in Fujian Province(JAT190397)。
文摘A discrete subset S of a topological gyrogroup G with the identity 0 is said to be a suitable set for G if it generates a dense subgyrogroup of G and S∪{0}is closed in G.In this paper,it is proved that each countable Hausdorff topological gyrogroup has a suitable set;moreover,it is shown that each separable metrizable strongly topological gyrogroup has a suitable set.
基金supported by the Basic Science Research Program through the National Research Foundation of Korea(NRF),funded by the Ministry of Education(RS-2023-00248378 and NRF-2020R1A6A1A03043708).
文摘Objectives:Tamoxifen is a key drug that provides endocrine therapy for estrogen receptor(ER)α-positive breast cancer;however,resistance remains a significant clinical challenge.This study aims to investigate the molecular mechanisms of tamoxifen resistance in ERα-positive breast cancer,with particular focus on the role of SET Domain Containing 1A(SETD1A)-driven forkhead box A2(FOXA2)as a key regulator of this resistance.Methods:FOXA2 expression and its regulation by SETD1A were assessed via(quantitative polymerase chain reaction),western blotting,transcriptome profiling,and chromatin immunoprecipitation analyses.The effects of FOXA2 on cell proliferation,migration,invasion,and cancer stem cell traits were evaluated using small interfering RNA(siRNA)-mediated silencing.Clinical relevance was examined by analyzing patient datasets and tumor tissue microarrays.Results:FOXA2 expression was significantly elevated in tamoxifen-resistant(TamR)and ERα-negative breast cancer cells compared to that in ERα-positive MCF-7 cells,regardless of tamoxifen treatment or ERαdepletion.Transcriptome and chromatin immunoprecipitation analyses revealed that SETD1A,a histone methyltransferase,directly regulated FOXA2 expression.Functionally,FOXA2 knockdown inhibited the proliferation,migration,invasion,and cancer stem cell properties of TamR cells while restoring tamoxifen sensitivity.High FOXA2 expression was correlated with poor survival and reduced responsiveness to tamoxifen in patients with ER-positive breast cancer.Conclusion:Our findings identified FOXA2 as a key mediator of tamoxifen resistance regulated by SETD1A and suggested that targeting the SETD1A-FOXA2 axis may offer a novel strategy for overcoming endocrine resistance in breast cancer.
文摘The quality of cardiopulmonary resuscitation(CPR) significantly influences survival and neurological outcomes in patients with cardiac arrest(CA).Although mechanical chest compression devices and extracorporeal cardiopulmonary resuscitation(ECPR) have demonstrated some benefits,high-quality manual CPR remained the essential first step,particularly in resource-limited settings.In this study,we examined whether opportunities existed to improve manual CPR performance using preliminary data from our recent survey conducted in a province in western China.We aim to emphasize the importance of improving manual CPR quality before implementing advanced interventions.
基金Project supported by the National Basic Research Program of China (973 Program) (No. 2007CB714600)
文摘The non-dominated sorting genetic algorithm （NSGA） is improved with the controlled elitism and dynamic crowding distance. A novel multi-objective optimization algorithm is obtained for wind turbine blades. As an example, a 5 MW wind turbine blade design is presented by taking the maximum power coefficient and the minimum blade mass as the optimization objectives. The optimal results show that this algorithm has good performance in handling the multi-objective optimization of wind turbines, and it gives a Pareto-optimal solution set rather than the optimum solutions to the conventional multi objective optimization problems. The wind turbine blade optimization method presented in this paper provides a new and general algorithm for the multi-objective optimization of wind turbines.
基金Foundation item： Projects（61102106, 61102105） supported by the National Natural Science Foundation of China Project（2013M530148） supported by China Postdoctoral Science Foundation Project（HEUCF120806） supported by the Fundamental Research Funds for the Central Universities of China
文摘In order to solve discrete multi-objective optimization problems, a non-dominated sorting quantum particle swarm optimization (NSQPSO) based on non-dominated sorting and quantum particle swarm optimization is proposed, and the performance of the NSQPSO is evaluated through five classical benchmark functions. The quantum particle swarm optimization (QPSO) applies the quantum computing theory to particle swarm optimization, and thus has the advantages of both quantum computing theory and particle swarm optimization, so it has a faster convergence rate and a more accurate convergence value. Therefore, QPSO is used as the evolutionary method of the proposed NSQPSO. Also NSQPSO is used to solve cognitive radio spectrum allocation problem. The methods to complete spectrum allocation in previous literature only consider one objective, i.e. network utilization or fairness, but the proposed NSQPSO method, can consider both network utilization and fairness simultaneously through obtaining Pareto front solutions. Cognitive radio systems can select one solution from the Pareto front solutions according to the weight of network reward and fairness. If one weight is unit and the other is zero, then it becomes single objective optimization, so the proposed NSQPSO method has a much wider application range. The experimental research results show that the NSQPS can obtain the same non-dominated solutions as exhaustive search but takes much less time in small dimensions; while in large dimensions, where the problem cannot be solved by exhaustive search, the NSQPSO can still solve the problem, which proves the effectiveness of NSQPSO.
基金the Natural Science Key Foundation of Heilongjiang Province of China (No. ZJG0503) China-UK Sci-ence Network from Royal Society UK
文摘Through the transformation of hydraulic constraints into the objective functions associated with a water supply network rehabilitation problem, a non-dominated sorting Genetic Algorithm-II (NSGA-II) can be used to solve the altered multi-objective optimization model. The introduction of NSGA-II into water supply network optimal rehabilitation problem solves the conflict between one fitness value of standard genetic algorithm (SGA) and multi-objectives of rehabilitation problem. And the uncertainties brought by using weight coefficients or punish functions in conventional methods are controlled. And also by in-troduction of artificial inducement mutation (AIM) operation, the convergence speed of population is accelerated;this operation not only improves the convergence speed, but also improves the rationality and feasibility of solutions.
基金supported by National Natural Science Foundation of China(No.60474059)Hi-tech Research and Development Program of China(863 Program,No.2006AA04Z160).
文摘Vehicle routing problem in distribution(VRPD)is a widely used type of vehicle routing problem(VRP),which has been proved as NP-Hard,and it is usually modeled as single objective optimization problem when modeling.For multi-objective optimization model,most researches consider two objectives.A multi-objective mathematical model for VRP is proposed,which considers the number of vehicles used,the length of route and the time arrived at each client.Genetic algorithm is one of the most widely used algorithms to solve VRP.As a type of genetic algorithm(GA),non-dominated sorting in genetic algorithm-Ⅱ(NSGA-Ⅱ)also suffers from premature convergence and enclosure competition.In order to avoid these kinds of shortage,a greedy NSGA-Ⅱ(GNSGA-Ⅱ)is proposed for VRP problem.Greedy algorithm is implemented in generating the initial population,cross-over and mutation.All these procedures ensure that NSGA-Ⅱis prevented from premature convergence and refine the performance of NSGA-Ⅱat each step.In the distribution problem of a distribution center in Michigan,US,the GNSGA-Ⅱis compared with NSGA-Ⅱ.As a result,the GNSGA-Ⅱis the most efficient one and can get the most optimized solution to VRP problem.Also,in GNSGA-Ⅱ,premature convergence is better avoided and search efficiency has been improved sharply.
基金supported by the Styrelsen för Internationellt Utvecklingssamarbete.
文摘With the continuous increase of rapid urbanization and population growth,sustainable urban land-use planning is becoming a more complex and challenging task for urban planners and decision-makers.Multi-objective land-use allocation can be regarded as a complex spatial optimization problem that aims to achieve the possible trade-offs among multiple and conflicting objectives.This paper proposes an improved Non-dominated Sorting Biogeography-Based Optimization(NSBBO)algorithm for solving the multi-objective land-use allocation problem,in which maximum accessibility,maximum compactness,and maximum spatial integration were formulated as spatial objectives;and space syntax analysis was used to analyze the potential movement patterns in the new urban planning area of the city of Kigali,Rwanda.Efficient Non-dominated Sorting(ENS)algorithm and crossover operator were integrated into classical NSBBO to improve the quality of non-dominated solutions,and local search ability,and to accelerate the convergence speed of the algorithm.The results showed that the proposed NSBBO exhibited good optimal solutions with a high hypervolume index compared to the classical NSBBO.Furthermore,the proposed algorithm could generate optimal land use scenarios according to the preferred objectives,thus having the potential to support the decision-making of urban planners and stockholders in revising and updating the existing detailed master plan of land use.
文摘This paper uses the Butterfly Optimization Algorithm(BOA)with dominated sorting and crowding distance mechanisms to solve multi-objective optimization problems.There is also an improvement to the original version of BOA to alleviate its drawbacks before extending it into a multi-objective version.Due to better coverage and a well-distributed Pareto front,non-dominant rankings are applied to the modified BOA using the crowding distance strategy.Seven benchmark functions and eight real-world problems have been used to test the performance of multi-objective non-dominated advanced BOA(MONSBOA),including unconstrained,constrained,and real-world design multiple-objective,highly nonlinear constraint problems.Various performance metrics,such as Generational Distance(GD),Inverted Generational Distance(IGD),Maximum Spread(MS),and Spacing(S),have been used for performance comparison.It is demonstrated that the new MONSBOA algorithm is better than the compared algorithms in more than 80%occasions in solving problems with a variety of linear,nonlinear,continuous,and discrete characteristics based on the Pareto front when compared quantitatively.From all the analysis,it may be concluded that the suggested MONSBOA is capable of producing high-quality Pareto fronts with very competitive results with rapid convergence.
基金This research was supported by the Researchers Supporting Program(TUMAProject-2021-27)Almaarefa University,Riyadh,Saudi Arabia.
文摘In computer vision,convolutional neural networks have a wide range of uses.Images representmost of today’s data,so it’s important to know how to handle these large amounts of data efficiently.Convolutional neural networks have been shown to solve image processing problems effectively.However,when designing the network structure for a particular problem,you need to adjust the hyperparameters for higher accuracy.This technique is time consuming and requires a lot of work and domain knowledge.Designing a convolutional neural network architecture is a classic NP-hard optimization challenge.On the other hand,different datasets require different combinations of models or hyperparameters,which can be time consuming and inconvenient.Various approaches have been proposed to overcome this problem,such as grid search limited to low-dimensional space and queuing by random selection.To address this issue,we propose an evolutionary algorithm-based approach that dynamically enhances the structure of Convolution Neural Networks(CNNs)using optimized hyperparameters.This study proposes a method using Non-dominated sorted genetic algorithms(NSGA)to improve the hyperparameters of the CNN model.In addition,different types and parameter ranges of existing genetic algorithms are used.Acomparative study was conducted with various state-of-the-art methodologies and algorithms.Experiments have shown that our proposed approach is superior to previous methods in terms of classification accuracy,and the results are published in modern computing literature.
