Designing appropriate loss functions is critical to the success of supervised learning models.However,most conventional losses are fixed and manually designed,making them suboptimal for diverse and dynamic learning sc...Designing appropriate loss functions is critical to the success of supervised learning models.However,most conventional losses are fixed and manually designed,making them suboptimal for diverse and dynamic learning scenarios.In this work,we propose an Adaptive Meta-Loss Network(Adaptive-MLN)that learns to generate taskagnostic loss functions tailored to evolving classification problems.Unlike traditional methods that rely on static objectives,Adaptive-MLN treats the loss function itself as a trainable component,parameterized by a shallow neural network.To enable flexible,gradient-free optimization,we introduce a hybrid evolutionary approach that combines GeneticAlgorithms(GA)for global exploration and Evolution Strategies(ES)for local refinement.This co-evolutionary process dynamically adjusts the loss landscape,improvingmodel generalization without relying on analytic gradients or handcrafted heuristics.Experimental evaluations on synthetic tasks and the CIFAR-10 andMNIST datasets demonstrate that our approach consistently outperforms standard losses such as Cross-Entropy and Mean Squared Error in terms of accuracy,convergence,and adaptability.展开更多
In recent years,feature selection(FS)optimization of high-dimensional gene expression data has become one of the most promising approaches for cancer prediction and classification.This work reviews FS and classificati...In recent years,feature selection(FS)optimization of high-dimensional gene expression data has become one of the most promising approaches for cancer prediction and classification.This work reviews FS and classification methods that utilize evolutionary algorithms(EAs)for gene expression profiles in cancer or medical applications based on research motivations,challenges,and recommendations.Relevant studies were retrieved from four major academic databases-IEEE,Scopus,Springer,and ScienceDirect-using the keywords‘cancer classification’,‘optimization’,‘FS’,and‘gene expression profile’.A total of 67 papers were finally selected with key advancements identified as follows:(1)The majority of papers(44.8%)focused on developing algorithms and models for FS and classification.(2)The second category encompassed studies on biomarker identification by EAs,including 20 papers(30%).(3)The third category comprised works that applied FS to cancer data for decision support system purposes,addressing high-dimensional data and the formulation of chromosome length.These studies accounted for 12%of the total number of studies.(4)The remaining three papers(4.5%)were reviews and surveys focusing on models and developments in prediction and classification optimization for cancer classification under current technical conditions.This review highlights the importance of optimizing FS in EAs to manage high-dimensional data effectively.Despite recent advancements,significant limitations remain:the dynamic formulation of chromosome length remains an underexplored area.Thus,further research is needed on dynamic-length chromosome techniques for more sophisticated biomarker gene selection techniques.The findings suggest that further advancements in dynamic chromosome length formulations and adaptive algorithms could enhance cancer classification accuracy and efficiency.展开更多
In recent years,the development of new types of nuclear reactors,such as transportable,marine,and space reactors,has presented new challenges for the optimization of reactor radiation-shielding design.Shielding struct...In recent years,the development of new types of nuclear reactors,such as transportable,marine,and space reactors,has presented new challenges for the optimization of reactor radiation-shielding design.Shielding structures typically need to be lightweight,miniaturized,and radiation-protected,which is a multi-parameter and multi-objective optimization problem.The conventional multi-objective(two or three objectives)optimization method for radiation-shielding design exhibits limitations for a number of optimization objectives and variable parameters,as well as a deficiency in achieving a global optimal solution,thereby failing to meet the requirements of shielding optimization for newly developed reactors.In this study,genetic and artificial bee-colony algorithms are combined with a reference-point-selection strategy and applied to the many-objective(having four or more objectives)optimal design of reactor radiation shielding.To validate the reliability of the methods,an optimization simulation is conducted on three-dimensional shielding structures and another complicated shielding-optimization problem.The numerical results demonstrate that the proposed algorithms outperform conventional shielding-design methods in terms of optimization performance,and they exhibit their reliability in practical engineering problems.The many-objective optimization algorithms developed in this study are proven to efficiently and consistently search for Pareto-front shielding schemes.Therefore,the algorithms proposed in this study offer novel insights into improving the shielding-design performance and shielding quality of new reactor types.展开更多
Nonlinear equations systems(NESs)are widely used in real-world problems and they are difficult to solve due to their nonlinearity and multiple roots.Evolutionary algorithms(EAs)are one of the methods for solving NESs,...Nonlinear equations systems(NESs)are widely used in real-world problems and they are difficult to solve due to their nonlinearity and multiple roots.Evolutionary algorithms(EAs)are one of the methods for solving NESs,given their global search capabilities and ability to locate multiple roots of a NES simultaneously within one run.Currently,the majority of research on using EAs to solve NESs focuses on transformation techniques and improving the performance of the used EAs.By contrast,problem domain knowledge of NESs is investigated in this study,where we propose the incorporation of a variable reduction strategy(VRS)into EAs to solve NESs.The VRS makes full use of the systems of expressing a NES and uses some variables(i.e.,core variable)to represent other variables(i.e.,reduced variables)through variable relationships that exist in the equation systems.It enables the reduction of partial variables and equations and shrinks the decision space,thereby reducing the complexity of the problem and improving the search efficiency of the EAs.To test the effectiveness of VRS in dealing with NESs,this paper mainly integrates the VRS into two existing state-of-the-art EA methods(i.e.,MONES and DR-JADE)according to the integration framework of the VRS and EA,respectively.Experimental results show that,with the assistance of the VRS,the EA methods can produce better results than the original methods and other compared methods.Furthermore,extensive experiments regarding the influence of different reduction schemes and EAs substantiate that a better EA for solving a NES with more reduced variables tends to provide better performance.展开更多
Evolutionary computation is a kind of adaptive non--numerical computation method which is designed tosimulate evolution of nature. In this paper, evolutionary algorithm behavior is described in terms of theconstructio...Evolutionary computation is a kind of adaptive non--numerical computation method which is designed tosimulate evolution of nature. In this paper, evolutionary algorithm behavior is described in terms of theconstruction and evolution of the sampling distributions over the space of candidate solutions. Iterativeconstruction of the sampling distributions is based on the idea of the global random search of generationalmethods. Under this frame, propontional selection is characterized as a gobal search operator, and recombination is characerized as the search process that exploits similarities. It is shown-that by properly constraining the search breadth of recombination operators, weak convergence of evolutionary algorithms to aglobal optimum can be ensured.展开更多
Multi-objective Evolutionary Algorithm (MOEA) is becoming a hot research area and quite a few aspects of MOEAs have been studied and discussed. However there are still few literatures discussing the roles of search an...Multi-objective Evolutionary Algorithm (MOEA) is becoming a hot research area and quite a few aspects of MOEAs have been studied and discussed. However there are still few literatures discussing the roles of search and selection operators in MOEAs. This paper studied their roles by solving a case of discrete Multi-objective Optimization Problem (MOP): Multi-objective TSP with a new MOEA. In the new MOEA, We adopt an efficient search operator, which has the properties of both crossover and mutation, to generate the new individuals and chose two selection operators: Family Competition and Population Competition with probabilities to realize selection. The simulation experiments showed that this new MOEA could get good uniform solutions representing the Pareto Front and outperformed SPEA in almost every simulation run on this problem. Furthermore, we analyzed its convergence property using finite Markov chain and proved that it could converge to Pareto Front with probability 1. We also find that the convergence property of MOEAs has much relationship with search and selection operators.展开更多
Multi-Objective Evolutionary Algorithms(MOEAs)have significantly advanced the domain of MultiObjective Optimization(MOO),facilitating solutions for complex problems with multiple conflicting objectives.This review exp...Multi-Objective Evolutionary Algorithms(MOEAs)have significantly advanced the domain of MultiObjective Optimization(MOO),facilitating solutions for complex problems with multiple conflicting objectives.This review explores the historical development of MOEAs,beginning with foundational concepts in multi-objective optimization,basic types of MOEAs,and the evolution of Pareto-based selection and niching methods.Further advancements,including decom-position-based approaches and hybrid algorithms,are discussed.Applications are analyzed in established domains such as engineering and economics,as well as in emerging fields like advanced analytics and machine learning.The significance of MOEAs in addressing real-world problems is emphasized,highlighting their role in facilitating informed decision-making.Finally,the development trajectory of MOEAs is compared with evolutionary processes,offering insights into their progress and future potential.展开更多
Convolutional neural networks(CNNs)exhibit superior performance in image feature extraction,making them extensively used in the area of traffic sign recognition.However,the design of existing traffic sign recognition ...Convolutional neural networks(CNNs)exhibit superior performance in image feature extraction,making them extensively used in the area of traffic sign recognition.However,the design of existing traffic sign recognition algorithms often relies on expert knowledge to enhance the image feature extraction networks,necessitating image preprocessing and model parameter tuning.This increases the complexity of the model design process.This study introduces an evolutionary neural architecture search(ENAS)algorithm for the automatic design of neural network models tailored for traffic sign recognition.By integrating the construction parameters of residual network(ResNet)into evolutionary algorithms(EAs),we automatically generate lightweight networks for traffic sign recognition,utilizing blocks as the fundamental building units.Experimental evaluations on the German traffic sign recognition benchmark(GTSRB)dataset reveal that the algorithm attains a recognition accuracy of 99.32%,with a mere 2.8×10^(6)parameters.Experimental results comparing the proposed method with other traffic sign recognition algorithms demonstrate that the method can more efficiently discover neural network architectures,significantly reducing the number of network parameters while maintaining recognition accuracy.展开更多
Rolling element bearing is the most common machine element in rotating machinery.An extended life is among the foremost imperative standards in the optimal design of rolling element bearings,which confide on the fatig...Rolling element bearing is the most common machine element in rotating machinery.An extended life is among the foremost imperative standards in the optimal design of rolling element bearings,which confide on the fatigue failure,wear,and thermal conditions of bearings.To fill the gap,in the current work,all three objectives of a tapered roller bearing have been innovatively considered respectively,which are the dynamic capacity,elasto-hydrodynamic lubrication(EHL)minimum film⁃thickness,and maximum bearing temperature.These objective function formulations are presented,associated design variables are identified,and constraints are discussed.To solve complex non⁃linear constrained optimization formulations,a best⁃practice design procedure was investigated using the Artificial Bee Colony(ABC)algorithms.A sensitivity analysis of several geometric design variables was conducted to observe the difference in all three objectives.An excellent enhancement was found in the bearing designs that have been optimized as compared with bearing standards and previously published works.The present study will definitely add to the present experience based design followed in bearing industries to save time and obtain assessment of bearing performance before manufacturing.To verify the improvement,an experimental investigation is worthwhile conducting.展开更多
In a wide range of engineering applications,complex constrained multi-objective optimization problems(CMOPs)present significant challenges,as the complexity of constraints often hampers algorithmic convergence and red...In a wide range of engineering applications,complex constrained multi-objective optimization problems(CMOPs)present significant challenges,as the complexity of constraints often hampers algorithmic convergence and reduces population diversity.To address these challenges,we propose a novel algorithm named Constraint IntensityDriven Evolutionary Multitasking(CIDEMT),which employs a two-stage,tri-task framework to dynamically integrates problem structure and knowledge transfer.In the first stage,three cooperative tasks are designed to explore the Constrained Pareto Front(CPF),the Unconstrained Pareto Front(UPF),and theε-relaxed constraint boundary,respectively.A CPF-UPF relationship classifier is employed to construct a problem-type-aware evolutionary strategy pool.At the end of the first stage,each task selects strategies from this strategy pool based on the specific type of problem,thereby guiding the subsequent evolutionary process.In the second stage,while each task continues to evolve,aτ-driven knowledge transfer mechanism is introduced to selectively incorporate effective solutions across tasks.enhancing the convergence and feasibility of the main task.Extensive experiments conducted on 32 benchmark problems from three test suites(LIRCMOP,DASCMOP,and DOC)demonstrate that CIDEMT achieves the best Inverted Generational Distance(IGD)values on 24 problems and the best Hypervolume values(HV)on 22 problems.Furthermore,CIDEMT significantly outperforms six state-of-the-art constrained multi-objective evolutionary algorithms(CMOEAs).These results confirm CIDEMT’s superiority in promoting convergence,diversity,and robustness in solving complex CMOPs.展开更多
Community detection is one of the most fundamental applications in understanding the structure of complicated networks.Furthermore,it is an important approach to identifying closely linked clusters of nodes that may r...Community detection is one of the most fundamental applications in understanding the structure of complicated networks.Furthermore,it is an important approach to identifying closely linked clusters of nodes that may represent underlying patterns and relationships.Networking structures are highly sensitive in social networks,requiring advanced techniques to accurately identify the structure of these communities.Most conventional algorithms for detecting communities perform inadequately with complicated networks.In addition,they miss out on accurately identifying clusters.Since single-objective optimization cannot always generate accurate and comprehensive results,as multi-objective optimization can.Therefore,we utilized two objective functions that enable strong connections between communities and weak connections between them.In this study,we utilized the intra function,which has proven effective in state-of-the-art research studies.We proposed a new inter-function that has demonstrated its effectiveness by making the objective of detecting external connections between communities is to make them more distinct and sparse.Furthermore,we proposed a Multi-Objective community strength enhancement algorithm(MOCSE).The proposed algorithm is based on the framework of the Multi-Objective Evolutionary Algorithm with Decomposition(MOEA/D),integrated with a new heuristic mutation strategy,community strength enhancement(CSE).The results demonstrate that the model is effective in accurately identifying community structures while also being computationally efficient.The performance measures used to evaluate the MOEA/D algorithm in our work are normalized mutual information(NMI)and modularity(Q).It was tested using five state-of-the-art algorithms on social networks,comprising real datasets(Zachary,Dolphin,Football,Krebs,SFI,Jazz,and Netscience),as well as twenty synthetic datasets.These results provide the robustness and practical value of the proposed algorithm in multi-objective community identification.展开更多
Community detection is a fundamental problem in network analysis for identifying densely connected node clusters,with successful applications in diverse fields like social networks,recommendation systems,biology,and c...Community detection is a fundamental problem in network analysis for identifying densely connected node clusters,with successful applications in diverse fields like social networks,recommendation systems,biology,and cyberattack detection.Overlapping community detection refers to the case of a node belonging to multiple communities simultaneously,which is a much more meaningful and challenging task.Graph representation learning with Evolutionary Computation has been studied well in overlapping community detection to deal with complex network structures and characteristics.However,most of them focus on searching the entire solution space,which can be inefficient and lead to inadequate results.To overcome the problem,a structural feature node extraction method is first proposed that can effectively map a network into a structural embedding space.Thus,nodes within the network are classified into hierarchical levels based on their structural feature strength,and only nodes with relatively high strength are considered in subsequent search steps to reduce the search space.Then,a maximal-clique representation method is employed on the given vertex set to identify overlapping nodes.A hybrid clique-based multi-objective evolutionary algorithmwith decomposition method is designed to address cliques and the remaining unexplored nodes separately.The number of communities generated with this allocation method is closer to the actual partition count with high division quality.Experimental results on nine usually used real-world networks,five synthetic networks,and two large-scale networks demonstrate the effectiveness of the proposed methodology in terms of community quality and algorithmic efficiency,compared to traditional,MOEA-based,and graph embedding-based community detection algorithms.展开更多
In this paper, we conduct research on the mechanical automation technology based on the evolutionary algorithms and artifi cialintelligence theory. Intelligent control theory after 30 years of development has made gra...In this paper, we conduct research on the mechanical automation technology based on the evolutionary algorithms and artifi cialintelligence theory. Intelligent control theory after 30 years of development has made gratifying achievements. But intelligent control has notyet formed a complete and systematic theory, based on the analysis, design, and there are many important problems in the practical application.Intelligent information processing is the use of some of the experience and knowledge of information, and the combination of that upper andlower knowledge information processing method. It is expected to solve the problem of insufficient information of pathology, computationcomplexity and the problem of real-time requirements, using the mathematical model is diffi cult to describe the nonlinear problem, etc. Underthis basis, this paper proposes the new mechanical automation technology based on the evolutionary algorithms and artifi cial intelligence theoryto propose the new perspective of dealing with the related challenges.展开更多
Flexible job shop scheduling problems(FJSP)have received much attention from academia and industry for many years.Due to their exponential complexity,swarm intelligence(SI)and evolutionary algorithms(EA)are developed,...Flexible job shop scheduling problems(FJSP)have received much attention from academia and industry for many years.Due to their exponential complexity,swarm intelligence(SI)and evolutionary algorithms(EA)are developed,employed and improved for solving them.More than 60%of the publications are related to SI and EA.This paper intents to give a comprehensive literature review of SI and EA for solving FJSP.First,the mathematical model of FJSP is presented and the constraints in applications are summarized.Then,the encoding and decoding strategies for connecting the problem and algorithms are reviewed.The strategies for initializing algorithms?population and local search operators for improving convergence performance are summarized.Next,one classical hybrid genetic algorithm(GA)and one newest imperialist competitive algorithm(ICA)with variables neighborhood search(VNS)for solving FJSP are presented.Finally,we summarize,discus and analyze the status of SI and EA for solving FJSP and give insight into future research directions.展开更多
基金supported by the National Natural Science Foundation of China(NSFC)under Grant number:82171965.
文摘Designing appropriate loss functions is critical to the success of supervised learning models.However,most conventional losses are fixed and manually designed,making them suboptimal for diverse and dynamic learning scenarios.In this work,we propose an Adaptive Meta-Loss Network(Adaptive-MLN)that learns to generate taskagnostic loss functions tailored to evolving classification problems.Unlike traditional methods that rely on static objectives,Adaptive-MLN treats the loss function itself as a trainable component,parameterized by a shallow neural network.To enable flexible,gradient-free optimization,we introduce a hybrid evolutionary approach that combines GeneticAlgorithms(GA)for global exploration and Evolution Strategies(ES)for local refinement.This co-evolutionary process dynamically adjusts the loss landscape,improvingmodel generalization without relying on analytic gradients or handcrafted heuristics.Experimental evaluations on synthetic tasks and the CIFAR-10 andMNIST datasets demonstrate that our approach consistently outperforms standard losses such as Cross-Entropy and Mean Squared Error in terms of accuracy,convergence,and adaptability.
基金funded by the Ministry of Higher Education of Malaysia,grant number FRGS/1/2022/ICT02/UPSI/02/1.
文摘In recent years,feature selection(FS)optimization of high-dimensional gene expression data has become one of the most promising approaches for cancer prediction and classification.This work reviews FS and classification methods that utilize evolutionary algorithms(EAs)for gene expression profiles in cancer or medical applications based on research motivations,challenges,and recommendations.Relevant studies were retrieved from four major academic databases-IEEE,Scopus,Springer,and ScienceDirect-using the keywords‘cancer classification’,‘optimization’,‘FS’,and‘gene expression profile’.A total of 67 papers were finally selected with key advancements identified as follows:(1)The majority of papers(44.8%)focused on developing algorithms and models for FS and classification.(2)The second category encompassed studies on biomarker identification by EAs,including 20 papers(30%).(3)The third category comprised works that applied FS to cancer data for decision support system purposes,addressing high-dimensional data and the formulation of chromosome length.These studies accounted for 12%of the total number of studies.(4)The remaining three papers(4.5%)were reviews and surveys focusing on models and developments in prediction and classification optimization for cancer classification under current technical conditions.This review highlights the importance of optimizing FS in EAs to manage high-dimensional data effectively.Despite recent advancements,significant limitations remain:the dynamic formulation of chromosome length remains an underexplored area.Thus,further research is needed on dynamic-length chromosome techniques for more sophisticated biomarker gene selection techniques.The findings suggest that further advancements in dynamic chromosome length formulations and adaptive algorithms could enhance cancer classification accuracy and efficiency.
基金supported by the National Natural Science Foundation of China(Nos.12475174 and 12175101)Yue Lu Shan Center Industrial Innovation(No.2024YCII0108)。
文摘In recent years,the development of new types of nuclear reactors,such as transportable,marine,and space reactors,has presented new challenges for the optimization of reactor radiation-shielding design.Shielding structures typically need to be lightweight,miniaturized,and radiation-protected,which is a multi-parameter and multi-objective optimization problem.The conventional multi-objective(two or three objectives)optimization method for radiation-shielding design exhibits limitations for a number of optimization objectives and variable parameters,as well as a deficiency in achieving a global optimal solution,thereby failing to meet the requirements of shielding optimization for newly developed reactors.In this study,genetic and artificial bee-colony algorithms are combined with a reference-point-selection strategy and applied to the many-objective(having four or more objectives)optimal design of reactor radiation shielding.To validate the reliability of the methods,an optimization simulation is conducted on three-dimensional shielding structures and another complicated shielding-optimization problem.The numerical results demonstrate that the proposed algorithms outperform conventional shielding-design methods in terms of optimization performance,and they exhibit their reliability in practical engineering problems.The many-objective optimization algorithms developed in this study are proven to efficiently and consistently search for Pareto-front shielding schemes.Therefore,the algorithms proposed in this study offer novel insights into improving the shielding-design performance and shielding quality of new reactor types.
基金This work was supported by the National Natural Science Foundation of China(62073341)in part by the Natural Science Fund for Distinguished Young Scholars of Hunan Province(2019JJ20026).
文摘Nonlinear equations systems(NESs)are widely used in real-world problems and they are difficult to solve due to their nonlinearity and multiple roots.Evolutionary algorithms(EAs)are one of the methods for solving NESs,given their global search capabilities and ability to locate multiple roots of a NES simultaneously within one run.Currently,the majority of research on using EAs to solve NESs focuses on transformation techniques and improving the performance of the used EAs.By contrast,problem domain knowledge of NESs is investigated in this study,where we propose the incorporation of a variable reduction strategy(VRS)into EAs to solve NESs.The VRS makes full use of the systems of expressing a NES and uses some variables(i.e.,core variable)to represent other variables(i.e.,reduced variables)through variable relationships that exist in the equation systems.It enables the reduction of partial variables and equations and shrinks the decision space,thereby reducing the complexity of the problem and improving the search efficiency of the EAs.To test the effectiveness of VRS in dealing with NESs,this paper mainly integrates the VRS into two existing state-of-the-art EA methods(i.e.,MONES and DR-JADE)according to the integration framework of the VRS and EA,respectively.Experimental results show that,with the assistance of the VRS,the EA methods can produce better results than the original methods and other compared methods.Furthermore,extensive experiments regarding the influence of different reduction schemes and EAs substantiate that a better EA for solving a NES with more reduced variables tends to provide better performance.
文摘Evolutionary computation is a kind of adaptive non--numerical computation method which is designed tosimulate evolution of nature. In this paper, evolutionary algorithm behavior is described in terms of theconstruction and evolution of the sampling distributions over the space of candidate solutions. Iterativeconstruction of the sampling distributions is based on the idea of the global random search of generationalmethods. Under this frame, propontional selection is characterized as a gobal search operator, and recombination is characerized as the search process that exploits similarities. It is shown-that by properly constraining the search breadth of recombination operators, weak convergence of evolutionary algorithms to aglobal optimum can be ensured.
基金Supported by the National Natural Science Foundation of China(60133010,70071042,60073043)
文摘Multi-objective Evolutionary Algorithm (MOEA) is becoming a hot research area and quite a few aspects of MOEAs have been studied and discussed. However there are still few literatures discussing the roles of search and selection operators in MOEAs. This paper studied their roles by solving a case of discrete Multi-objective Optimization Problem (MOP): Multi-objective TSP with a new MOEA. In the new MOEA, We adopt an efficient search operator, which has the properties of both crossover and mutation, to generate the new individuals and chose two selection operators: Family Competition and Population Competition with probabilities to realize selection. The simulation experiments showed that this new MOEA could get good uniform solutions representing the Pareto Front and outperformed SPEA in almost every simulation run on this problem. Furthermore, we analyzed its convergence property using finite Markov chain and proved that it could converge to Pareto Front with probability 1. We also find that the convergence property of MOEAs has much relationship with search and selection operators.
文摘Multi-Objective Evolutionary Algorithms(MOEAs)have significantly advanced the domain of MultiObjective Optimization(MOO),facilitating solutions for complex problems with multiple conflicting objectives.This review explores the historical development of MOEAs,beginning with foundational concepts in multi-objective optimization,basic types of MOEAs,and the evolution of Pareto-based selection and niching methods.Further advancements,including decom-position-based approaches and hybrid algorithms,are discussed.Applications are analyzed in established domains such as engineering and economics,as well as in emerging fields like advanced analytics and machine learning.The significance of MOEAs in addressing real-world problems is emphasized,highlighting their role in facilitating informed decision-making.Finally,the development trajectory of MOEAs is compared with evolutionary processes,offering insights into their progress and future potential.
基金supported by the National Natural Science Foundation of China(No.62066041).
文摘Convolutional neural networks(CNNs)exhibit superior performance in image feature extraction,making them extensively used in the area of traffic sign recognition.However,the design of existing traffic sign recognition algorithms often relies on expert knowledge to enhance the image feature extraction networks,necessitating image preprocessing and model parameter tuning.This increases the complexity of the model design process.This study introduces an evolutionary neural architecture search(ENAS)algorithm for the automatic design of neural network models tailored for traffic sign recognition.By integrating the construction parameters of residual network(ResNet)into evolutionary algorithms(EAs),we automatically generate lightweight networks for traffic sign recognition,utilizing blocks as the fundamental building units.Experimental evaluations on the German traffic sign recognition benchmark(GTSRB)dataset reveal that the algorithm attains a recognition accuracy of 99.32%,with a mere 2.8×10^(6)parameters.Experimental results comparing the proposed method with other traffic sign recognition algorithms demonstrate that the method can more efficiently discover neural network architectures,significantly reducing the number of network parameters while maintaining recognition accuracy.
文摘Rolling element bearing is the most common machine element in rotating machinery.An extended life is among the foremost imperative standards in the optimal design of rolling element bearings,which confide on the fatigue failure,wear,and thermal conditions of bearings.To fill the gap,in the current work,all three objectives of a tapered roller bearing have been innovatively considered respectively,which are the dynamic capacity,elasto-hydrodynamic lubrication(EHL)minimum film⁃thickness,and maximum bearing temperature.These objective function formulations are presented,associated design variables are identified,and constraints are discussed.To solve complex non⁃linear constrained optimization formulations,a best⁃practice design procedure was investigated using the Artificial Bee Colony(ABC)algorithms.A sensitivity analysis of several geometric design variables was conducted to observe the difference in all three objectives.An excellent enhancement was found in the bearing designs that have been optimized as compared with bearing standards and previously published works.The present study will definitely add to the present experience based design followed in bearing industries to save time and obtain assessment of bearing performance before manufacturing.To verify the improvement,an experimental investigation is worthwhile conducting.
基金supported by the National Natural Science Foundation of China under Grant No.61972040the Science and Technology Research and Development Project funded by China Railway Material Trade Group Luban Company.
文摘In a wide range of engineering applications,complex constrained multi-objective optimization problems(CMOPs)present significant challenges,as the complexity of constraints often hampers algorithmic convergence and reduces population diversity.To address these challenges,we propose a novel algorithm named Constraint IntensityDriven Evolutionary Multitasking(CIDEMT),which employs a two-stage,tri-task framework to dynamically integrates problem structure and knowledge transfer.In the first stage,three cooperative tasks are designed to explore the Constrained Pareto Front(CPF),the Unconstrained Pareto Front(UPF),and theε-relaxed constraint boundary,respectively.A CPF-UPF relationship classifier is employed to construct a problem-type-aware evolutionary strategy pool.At the end of the first stage,each task selects strategies from this strategy pool based on the specific type of problem,thereby guiding the subsequent evolutionary process.In the second stage,while each task continues to evolve,aτ-driven knowledge transfer mechanism is introduced to selectively incorporate effective solutions across tasks.enhancing the convergence and feasibility of the main task.Extensive experiments conducted on 32 benchmark problems from three test suites(LIRCMOP,DASCMOP,and DOC)demonstrate that CIDEMT achieves the best Inverted Generational Distance(IGD)values on 24 problems and the best Hypervolume values(HV)on 22 problems.Furthermore,CIDEMT significantly outperforms six state-of-the-art constrained multi-objective evolutionary algorithms(CMOEAs).These results confirm CIDEMT’s superiority in promoting convergence,diversity,and robustness in solving complex CMOPs.
文摘Community detection is one of the most fundamental applications in understanding the structure of complicated networks.Furthermore,it is an important approach to identifying closely linked clusters of nodes that may represent underlying patterns and relationships.Networking structures are highly sensitive in social networks,requiring advanced techniques to accurately identify the structure of these communities.Most conventional algorithms for detecting communities perform inadequately with complicated networks.In addition,they miss out on accurately identifying clusters.Since single-objective optimization cannot always generate accurate and comprehensive results,as multi-objective optimization can.Therefore,we utilized two objective functions that enable strong connections between communities and weak connections between them.In this study,we utilized the intra function,which has proven effective in state-of-the-art research studies.We proposed a new inter-function that has demonstrated its effectiveness by making the objective of detecting external connections between communities is to make them more distinct and sparse.Furthermore,we proposed a Multi-Objective community strength enhancement algorithm(MOCSE).The proposed algorithm is based on the framework of the Multi-Objective Evolutionary Algorithm with Decomposition(MOEA/D),integrated with a new heuristic mutation strategy,community strength enhancement(CSE).The results demonstrate that the model is effective in accurately identifying community structures while also being computationally efficient.The performance measures used to evaluate the MOEA/D algorithm in our work are normalized mutual information(NMI)and modularity(Q).It was tested using five state-of-the-art algorithms on social networks,comprising real datasets(Zachary,Dolphin,Football,Krebs,SFI,Jazz,and Netscience),as well as twenty synthetic datasets.These results provide the robustness and practical value of the proposed algorithm in multi-objective community identification.
基金supported in part by the National Natural Science Foundation of China under Grants 62473176,62073155,62002137,62106088,and 62206113National Key Laboratory of Ship Structural Safety underGrant 450324300the Postgraduate Research&Practice Innovation Programof Jiangsu Province under Grant KYCX24_2642.
文摘Community detection is a fundamental problem in network analysis for identifying densely connected node clusters,with successful applications in diverse fields like social networks,recommendation systems,biology,and cyberattack detection.Overlapping community detection refers to the case of a node belonging to multiple communities simultaneously,which is a much more meaningful and challenging task.Graph representation learning with Evolutionary Computation has been studied well in overlapping community detection to deal with complex network structures and characteristics.However,most of them focus on searching the entire solution space,which can be inefficient and lead to inadequate results.To overcome the problem,a structural feature node extraction method is first proposed that can effectively map a network into a structural embedding space.Thus,nodes within the network are classified into hierarchical levels based on their structural feature strength,and only nodes with relatively high strength are considered in subsequent search steps to reduce the search space.Then,a maximal-clique representation method is employed on the given vertex set to identify overlapping nodes.A hybrid clique-based multi-objective evolutionary algorithmwith decomposition method is designed to address cliques and the remaining unexplored nodes separately.The number of communities generated with this allocation method is closer to the actual partition count with high division quality.Experimental results on nine usually used real-world networks,five synthetic networks,and two large-scale networks demonstrate the effectiveness of the proposed methodology in terms of community quality and algorithmic efficiency,compared to traditional,MOEA-based,and graph embedding-based community detection algorithms.
文摘In this paper, we conduct research on the mechanical automation technology based on the evolutionary algorithms and artifi cialintelligence theory. Intelligent control theory after 30 years of development has made gratifying achievements. But intelligent control has notyet formed a complete and systematic theory, based on the analysis, design, and there are many important problems in the practical application.Intelligent information processing is the use of some of the experience and knowledge of information, and the combination of that upper andlower knowledge information processing method. It is expected to solve the problem of insufficient information of pathology, computationcomplexity and the problem of real-time requirements, using the mathematical model is diffi cult to describe the nonlinear problem, etc. Underthis basis, this paper proposes the new mechanical automation technology based on the evolutionary algorithms and artifi cial intelligence theoryto propose the new perspective of dealing with the related challenges.
基金supported in part by the National Natural Science Foundation of China(61603169,61773192,61803192)in part by the funding from Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technologyin part by Singapore National Research Foundation(NRF-RSS2016-004)
文摘Flexible job shop scheduling problems(FJSP)have received much attention from academia and industry for many years.Due to their exponential complexity,swarm intelligence(SI)and evolutionary algorithms(EA)are developed,employed and improved for solving them.More than 60%of the publications are related to SI and EA.This paper intents to give a comprehensive literature review of SI and EA for solving FJSP.First,the mathematical model of FJSP is presented and the constraints in applications are summarized.Then,the encoding and decoding strategies for connecting the problem and algorithms are reviewed.The strategies for initializing algorithms?population and local search operators for improving convergence performance are summarized.Next,one classical hybrid genetic algorithm(GA)and one newest imperialist competitive algorithm(ICA)with variables neighborhood search(VNS)for solving FJSP are presented.Finally,we summarize,discus and analyze the status of SI and EA for solving FJSP and give insight into future research directions.
