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
This study concentrates of the new generation of the agile （AEOS）. AEOS is a key study object on management problems earth observation satellite in many countries because of its many advantages over non-agile satell...This study concentrates of the new generation of the agile （AEOS）. AEOS is a key study object on management problems earth observation satellite in many countries because of its many advantages over non-agile satellites. Hence, the mission planning and scheduling of AEOS is a popular research problem. This research investigates AEOS characteristics and establishes a mission planning model based on the working principle and constraints of AEOS as per analysis. To solve the scheduling issue of AEOS, several improved algorithms are developed. Simulation results suggest that these algorithms are effective.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
In this paper,a comparative study of the path planning problem using evolutionary algorithms,in comparison with classical methods such as the A*algorithm,is presented for a holonomic mobile robot.The configured naviga...In this paper,a comparative study of the path planning problem using evolutionary algorithms,in comparison with classical methods such as the A*algorithm,is presented for a holonomic mobile robot.The configured navigation system,which consists of the integration of sensors sources,map formatting,global and local path planners,and the base controller,aims to enable the robot to follow the shortest smooth path delicately.Grid-based mapping is used for scoring paths efficiently,allowing the determination of collision-free trajectories from the initial to the target position.This work considers the evolutionary algorithms,the mutated cuckoo optimization algorithm(MCOA)and the genetic algorithm(GA),as a global planner to find the shortest safe path among others.A non-uniform motion coefficient is introduced for MCOA in order to increase the performance of this algorithm.A series of experiments are accomplished and analyzed to confirm the performance of the global planner implemented on a holonomic mobile robot.The results of the experiments show the capacity of the planner framework with respect to the path planning problem under various obstacle layouts.展开更多
Surrogate models have shown to be effective in assisting evolutionary algorithms(EAs)for solving computationally expensive complex optimization problems.However,the effectiveness of the existing surrogate-assisted evo...Surrogate models have shown to be effective in assisting evolutionary algorithms(EAs)for solving computationally expensive complex optimization problems.However,the effectiveness of the existing surrogate-assisted evolutionary algorithms still needs to be improved.A data-driven evolutionary sampling optimization(DESO)framework is proposed,where at each generation it randomly employs one of two evolutionary sampling strategies,surrogate screening and surrogate local search based on historical data,to effectively balance global and local search.In DESO,the radial basis function(RBF)is used as the surrogate model in the sampling strategy,and different degrees of the evolutionary process are used to sample candidate points.The sampled points by sampling strategies are evaluated,and then added into the database for the updating surrogate model and population in the next sampling.To get the insight of DESO,extensive experiments and analysis of DESO have been performed.The proposed algorithm presents superior computational efficiency and robustness compared with five state-of-the-art algorithms on benchmark problems from 20 to 200 dimensions.Besides,DESO is applied to an airfoil design problem to show its effectiveness.展开更多
Since traditional fault tolerance methods of electronic systems are based on redundant fault tolerance technique,and their structures are fixed when circuits are designed,the self-adaptive ability is limited.In order ...Since traditional fault tolerance methods of electronic systems are based on redundant fault tolerance technique,and their structures are fixed when circuits are designed,the self-adaptive ability is limited.In order to solve these problems,a novel circuit self-adaptive design technique based on evolvable hardware(EHW)is proposed.It features robustness,self-organization and self-adaption.It can be adapted to a complex environment through dynamic configuration of the circuit.In this paper,the proposed technique simulated.The consumption of hardware resources and the number of convergence iterations researched.The effectiveness and superiority of the proposed technique are verified.The designed circuit has the ability of resistible redundant-state interference(RRSI).The proposed technique has a broad application prospect,and it has great significance.展开更多
Based on the theories of EA （Evolutionary Algorithm） and EHW （Evolvable Hardware）, we devise an EHW based software-hardware co designing platform ECDP, on which we provided standards for hardware system encoding a...Based on the theories of EA （Evolutionary Algorithm） and EHW （Evolvable Hardware）, we devise an EHW based software-hardware co designing platform ECDP, on which we provided standards for hardware system encoding and evolving operation designing, as well as circuit emulating tools. The major features of this system are： two layer-encoding of circuit structure, off-line evolving with software cmulation and the evolving of genetic program designing. With this system, we implemented the auto designing of sonic software-hardware systems, like the random number generator.展开更多
In order to characterizc large fluctuations of the financial markets and optimize financial portfolio, a new dynamic asset control strategy was proposed in this work. Firstly, a random process item with variable jump ...In order to characterizc large fluctuations of the financial markets and optimize financial portfolio, a new dynamic asset control strategy was proposed in this work. Firstly, a random process item with variable jump intensity was introduced to the existing discrete microstructure model to denote large price fluctuations. The nonparametric method of LEE was used for detecting jumps. Further, the extended Kalman filter and the maximum likelihood method were applied to discrete microstructure modeling and the estimation of two market potential variables： market excess demand and liquidity. At last, based on the estimated variables, an assets allocation strategy using evolutionary algorithm was designed to control the weight of each asset dynamically. Case studies on IBM Stock show that jumps with variable intensity are detected successfully, and the assets allocation strategy may effectively keep the total assets growth or prevent assets loss at the stochastic financial market.展开更多
In recent decades,significant advancements have been made in the rigorous runtime analysis of evolutionary algorithms(EAs).However,in the context of non-elitist EAs and the use of crossover,it is challenging to engage...In recent decades,significant advancements have been made in the rigorous runtime analysis of evolutionary algorithms(EAs).However,in the context of non-elitist EAs and the use of crossover,it is challenging to engage in any meaningful theoretical discussion due to the increasing complexity of the EA's population distribution as the EA runs.This paper aims to gain insight into the rigorous runtime analysis of the(μ,λ)EA with crossover,focusing on its optimization of the Jump test function,by investigating the population distribution during the optimization process.It is proposed that,under typical circumstances,the population distribution will first reach a stable and fully-diverged state before attaining the global optimum.Consequently,the optimization process is divided into two parts,based on whether the population distribution has reached this state.By investigating this state,the authors are able to provide a better upper bound on the runtime of the EA.Furthermore,a series of experiments were conducted to validate our theoretical results,which also offered insights into the impact of different parameters on this state.展开更多
The mutation operations and related control parameters play important roles in the performance of the differential evolution algorithm.Learning optimal policies for these strategies and parameters through reinforcemen...The mutation operations and related control parameters play important roles in the performance of the differential evolution algorithm.Learning optimal policies for these strategies and parameters through reinforcement learning is a hot topic.However,most of the current studies focus on either mutation strategy selection or the control parameters alone while the others keep fixed or self-adaptive,resulting in deteriorated performances.To address this gap,this paper proposes a framework for the joint adaptation of mutation strategies and related control parameters based on deep reinforcement learning.In this method,the distributed proximal policy optimization algorithm is employed to train the agents to dynamically select the optimal combination of mutation strategies and control parameters.To enhance the agent’s learning of the optimal policy,information derived from fitness landscape analysis is incorporated into the state representations.The training is conducted on the black-box optimization benchmark test problems,which are capable of generating large-scale test instances.Numerical results on the new problems from CEC2013 and CEC2017 test suites,and the real-world application of rover trajectory planning demonstrate that the proposed approach achieves competitive performance compared to state-of-the-art methods.The adaptation behavior and the contribution of learning are also thoroughly analyzed.展开更多
This paper presents a new method to reduce the distribution system loss by feeder reconfiguration.This new method combines self-adaptive particle swarm optimization(SAPSO) with shuffled frog-leaping algorithm(SFLA) in...This paper presents a new method to reduce the distribution system loss by feeder reconfiguration.This new method combines self-adaptive particle swarm optimization(SAPSO) with shuffled frog-leaping algorithm(SFLA) in an attempt to find the global optimal solutions for the distribution feeder reconfiguration(DFR).In PSO algorithm,appropriate adjustment of the parameters is cumbersome and usually requires a lot of time and effort.Thus,a self-adaptive framework is proposed to improve the robustness of PSO.In SAPSO the learning factors of PSO coevolve with the particles.SFLA is combined with the SAPSO algorithm to improve its performance.The proposed algorithm is tested on two distribution test networks.The results of simulation show that the proposed algorithm is very powerful and guarantees to obtain the global optimization in minimum time.展开更多
Most of works on the time complexity analysis of evolutionary algorithms havealways focused on some artificial binary problems.The time complexity of the algorithms forcombinatorial optimisation has not been well unde...Most of works on the time complexity analysis of evolutionary algorithms havealways focused on some artificial binary problems.The time complexity of the algorithms forcombinatorial optimisation has not been well understood.This paper considers the time complexity ofan evolutionary algorithm for a classical combinatorial optimisation problem,to find the maximumcardinality matching in a graph.It is shown that the evolutionary algorithm can produce a matchingwith nearly maximum cardinality in average polynomial time.展开更多
Portfolio optimization is a classical and important problem in the field of asset management,which aims to achieve a trade-off between profit and risk.Previous portfolio optimization models use traditional risk measur...Portfolio optimization is a classical and important problem in the field of asset management,which aims to achieve a trade-off between profit and risk.Previous portfolio optimization models use traditional risk measurements such as variance,which symmetrically delineate both positive and negative sides and are not practical and stable.In this paper,a new model with cardinality constraints is first proposed,in which the idiosyncratic volatility factor is used to replace traditional risk measurements and can capture the risks of the portfolio in a more accurate way.The new model has practical constraints which involve the sparsity and irregularity of variables and make it challenging to be solved by traditional Multi-Objective Evolutionary Algorithms(MOEAs).To solve the model,a Learning-Guided Evolutionary Algorithm based on I_(ϵ+)indicator(I_(ϵ+)LGEA)is developed.In I_(ϵ+)LGEA,the I_(ϵ+)indicator is incorporated into the initialization and genetic operators to guarantee the sparsity of solutions and can help improve the convergence of the algorithm.And a new constraint-handling method based on I_(ϵ+)indicator is also adopted to ensure the feasibility of solutions.The experimental results on five portfolio trading datasets including up to 1226 assets show that I_(ϵ+)LGEA outperforms some state-of-the-art MOEAs in most cases.展开更多
Evolutionary computation （EC） is one of the fastest growing areas in computer science that solves intractable optimization problems by emulating biologic evolution and organizational behaviors in nature. To de- sign...Evolutionary computation （EC） is one of the fastest growing areas in computer science that solves intractable optimization problems by emulating biologic evolution and organizational behaviors in nature. To de- sign an EC algorithm, one needs to determine a set of algorithmic configurations like operator selections and parameter settings. How to design an effective and ef- ficient adaptation scheme for adjusting the configura- tions of EC algorithms has become a significant and promising research topic in the EC research community. This paper intends to provide a comprehensive survey on this rapidly growing field. We present a classification of adaptive EC （AEC） algorithms from the perspective of how an adaptation scheme is designed, involving the adaptation objects, adaptation evidences, and adapta- tion methods. In particular, by analyzing tile popula- tion distribution characteristics of EC algorithms, we discuss why and how the evolutionary state information of EC can be estimated and utilized for designing ef- fective EC adaptation schemes. Two AEC algorithms using the idea of evolutionary state estimation, includ- ing the clustering-based adaptive genetic algorithm and the adaptive particle swarm optimization algorithm are presented in detail. Some potential directions for the re- search of AECs are also discussed in this paper.展开更多
Wastewater treatment plays a crucial role in alleviating water shortages and protecting the environment from pollution.Due to the strong time variabilities and complex nonlinearities within wastewater treatment system...Wastewater treatment plays a crucial role in alleviating water shortages and protecting the environment from pollution.Due to the strong time variabilities and complex nonlinearities within wastewater treatment systems,devising an efficient optimal controller to reduce energy consumption while ensuring effluent quality is still a bottleneck that needs to be addressed.In this paper,in order to comprehensively consider different needs of the wastewater treatment process(WTTP),a two-objective model is to consider a scope,in which minimizing energy consumption and guaranteeing effluent quality are both considered to improve wastewater treatment efficiency.To efficiently solve the model functions,a grid-based dynamic multi-objective evolutionary decomposition algorithm,namely GD-MOEA/D,is designed.A GD-MOEA/D-based intelligent optimal controller(GD-MOEA/D-IOC)is devised to achieve tracking control of the main operating variables of the WTTP.Finally,the benchmark simulation model No.1(BSM1)is applied to verify the validity of the proposed approach.The experimental results demonstrate that the constructed models can catch the dynamics of WWTP accurately.Moreover,GD-MOEA/D has better optimization ability in solving the designed models.GD-MOEA/D-IOC can achieve a significant improvement in terms of reducing energy consumption and improving effluent quality.Therefore,the designed multi-objective intelligent optimal control method for WWTP has great potential to be applied to practical engineering since it can easily achieve a highly intelligent control in WTTP.展开更多
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(7127106671171065+1 种基金71202168)the Natural Science Foundation of Heilongjiang Province(GC13D506)
文摘This study concentrates of the new generation of the agile （AEOS）. AEOS is a key study object on management problems earth observation satellite in many countries because of its many advantages over non-agile satellites. Hence, the mission planning and scheduling of AEOS is a popular research problem. This research investigates AEOS characteristics and establishes a mission planning model based on the working principle and constraints of AEOS as per analysis. To solve the scheduling issue of AEOS, several improved algorithms are developed. Simulation results suggest that these algorithms are effective.
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In this paper,a comparative study of the path planning problem using evolutionary algorithms,in comparison with classical methods such as the A*algorithm,is presented for a holonomic mobile robot.The configured navigation system,which consists of the integration of sensors sources,map formatting,global and local path planners,and the base controller,aims to enable the robot to follow the shortest smooth path delicately.Grid-based mapping is used for scoring paths efficiently,allowing the determination of collision-free trajectories from the initial to the target position.This work considers the evolutionary algorithms,the mutated cuckoo optimization algorithm(MCOA)and the genetic algorithm(GA),as a global planner to find the shortest safe path among others.A non-uniform motion coefficient is introduced for MCOA in order to increase the performance of this algorithm.A series of experiments are accomplished and analyzed to confirm the performance of the global planner implemented on a holonomic mobile robot.The results of the experiments show the capacity of the planner framework with respect to the path planning problem under various obstacle layouts.
基金supported by the National Natural Science Foundation of China(62076225,62073300)the Natural Science Foundation for Distinguished Young Scholars of Hubei(2019CFA081)the Fundamental Research Funds for the Central Universities,China University of Geosciences(Wuhan)(CUGGC03).
文摘Surrogate models have shown to be effective in assisting evolutionary algorithms(EAs)for solving computationally expensive complex optimization problems.However,the effectiveness of the existing surrogate-assisted evolutionary algorithms still needs to be improved.A data-driven evolutionary sampling optimization(DESO)framework is proposed,where at each generation it randomly employs one of two evolutionary sampling strategies,surrogate screening and surrogate local search based on historical data,to effectively balance global and local search.In DESO,the radial basis function(RBF)is used as the surrogate model in the sampling strategy,and different degrees of the evolutionary process are used to sample candidate points.The sampled points by sampling strategies are evaluated,and then added into the database for the updating surrogate model and population in the next sampling.To get the insight of DESO,extensive experiments and analysis of DESO have been performed.The proposed algorithm presents superior computational efficiency and robustness compared with five state-of-the-art algorithms on benchmark problems from 20 to 200 dimensions.Besides,DESO is applied to an airfoil design problem to show its effectiveness.
基金This work was supported by National Natural Science Foundation of China(Nos.61271153 and 61372039).
文摘Since traditional fault tolerance methods of electronic systems are based on redundant fault tolerance technique,and their structures are fixed when circuits are designed,the self-adaptive ability is limited.In order to solve these problems,a novel circuit self-adaptive design technique based on evolvable hardware(EHW)is proposed.It features robustness,self-organization and self-adaption.It can be adapted to a complex environment through dynamic configuration of the circuit.In this paper,the proposed technique simulated.The consumption of hardware resources and the number of convergence iterations researched.The effectiveness and superiority of the proposed technique are verified.The designed circuit has the ability of resistible redundant-state interference(RRSI).The proposed technique has a broad application prospect,and it has great significance.
基金Supported by the National 863 Project(2002AA1Z1490)
文摘Based on the theories of EA （Evolutionary Algorithm） and EHW （Evolvable Hardware）, we devise an EHW based software-hardware co designing platform ECDP, on which we provided standards for hardware system encoding and evolving operation designing, as well as circuit emulating tools. The major features of this system are： two layer-encoding of circuit structure, off-line evolving with software cmulation and the evolving of genetic program designing. With this system, we implemented the auto designing of sonic software-hardware systems, like the random number generator.
基金Projects(71271215,71221061) supported by the National Natural Science Foundation of ChinaProject(2011DFA10440) supported by the International Science&Technology Cooperation Program of ChinaProject(CX2012B067) supported by Hunan Provincial Innovation Foundation for Postgraduate,China
文摘In order to characterizc large fluctuations of the financial markets and optimize financial portfolio, a new dynamic asset control strategy was proposed in this work. Firstly, a random process item with variable jump intensity was introduced to the existing discrete microstructure model to denote large price fluctuations. The nonparametric method of LEE was used for detecting jumps. Further, the extended Kalman filter and the maximum likelihood method were applied to discrete microstructure modeling and the estimation of two market potential variables： market excess demand and liquidity. At last, based on the estimated variables, an assets allocation strategy using evolutionary algorithm was designed to control the weight of each asset dynamically. Case studies on IBM Stock show that jumps with variable intensity are detected successfully, and the assets allocation strategy may effectively keep the total assets growth or prevent assets loss at the stochastic financial market.
文摘In recent decades,significant advancements have been made in the rigorous runtime analysis of evolutionary algorithms(EAs).However,in the context of non-elitist EAs and the use of crossover,it is challenging to engage in any meaningful theoretical discussion due to the increasing complexity of the EA's population distribution as the EA runs.This paper aims to gain insight into the rigorous runtime analysis of the(μ,λ)EA with crossover,focusing on its optimization of the Jump test function,by investigating the population distribution during the optimization process.It is proposed that,under typical circumstances,the population distribution will first reach a stable and fully-diverged state before attaining the global optimum.Consequently,the optimization process is divided into two parts,based on whether the population distribution has reached this state.By investigating this state,the authors are able to provide a better upper bound on the runtime of the EA.Furthermore,a series of experiments were conducted to validate our theoretical results,which also offered insights into the impact of different parameters on this state.
基金supported by the National Natural Science Foundation of China(No.52105244)the Open Foundation of the State Key Laboratory of Fluid Power and Mechatronic Systems(No.GZKF-202425)+1 种基金the Entrepreneurship and Innovation Support Plan of Chongqing for Returned Overseas Scholars(No.cx2023085)the Independent Research Project-Key Program from the State Key Laboratory of Mechanical Transmission for Advanced Equipment(No.SKLMT-ZZKT-2024Z09).
文摘The mutation operations and related control parameters play important roles in the performance of the differential evolution algorithm.Learning optimal policies for these strategies and parameters through reinforcement learning is a hot topic.However,most of the current studies focus on either mutation strategy selection or the control parameters alone while the others keep fixed or self-adaptive,resulting in deteriorated performances.To address this gap,this paper proposes a framework for the joint adaptation of mutation strategies and related control parameters based on deep reinforcement learning.In this method,the distributed proximal policy optimization algorithm is employed to train the agents to dynamically select the optimal combination of mutation strategies and control parameters.To enhance the agent’s learning of the optimal policy,information derived from fitness landscape analysis is incorporated into the state representations.The training is conducted on the black-box optimization benchmark test problems,which are capable of generating large-scale test instances.Numerical results on the new problems from CEC2013 and CEC2017 test suites,and the real-world application of rover trajectory planning demonstrate that the proposed approach achieves competitive performance compared to state-of-the-art methods.The adaptation behavior and the contribution of learning are also thoroughly analyzed.
文摘This paper presents a new method to reduce the distribution system loss by feeder reconfiguration.This new method combines self-adaptive particle swarm optimization(SAPSO) with shuffled frog-leaping algorithm(SFLA) in an attempt to find the global optimal solutions for the distribution feeder reconfiguration(DFR).In PSO algorithm,appropriate adjustment of the parameters is cumbersome and usually requires a lot of time and effort.Thus,a self-adaptive framework is proposed to improve the robustness of PSO.In SAPSO the learning factors of PSO coevolve with the particles.SFLA is combined with the SAPSO algorithm to improve its performance.The proposed algorithm is tested on two distribution test networks.The results of simulation show that the proposed algorithm is very powerful and guarantees to obtain the global optimization in minimum time.
基金supported by Engineering and Physical Sciences Research Council(GR/R52541/01)State Key Lab of Software Engineering at Wuhan University
文摘Most of works on the time complexity analysis of evolutionary algorithms havealways focused on some artificial binary problems.The time complexity of the algorithms forcombinatorial optimisation has not been well understood.This paper considers the time complexity ofan evolutionary algorithm for a classical combinatorial optimisation problem,to find the maximumcardinality matching in a graph.It is shown that the evolutionary algorithm can produce a matchingwith nearly maximum cardinality in average polynomial time.
基金This work was supported by the National Natural Science Foundation of China(Nos.62173258 and 61773296).
文摘Portfolio optimization is a classical and important problem in the field of asset management,which aims to achieve a trade-off between profit and risk.Previous portfolio optimization models use traditional risk measurements such as variance,which symmetrically delineate both positive and negative sides and are not practical and stable.In this paper,a new model with cardinality constraints is first proposed,in which the idiosyncratic volatility factor is used to replace traditional risk measurements and can capture the risks of the portfolio in a more accurate way.The new model has practical constraints which involve the sparsity and irregularity of variables and make it challenging to be solved by traditional Multi-Objective Evolutionary Algorithms(MOEAs).To solve the model,a Learning-Guided Evolutionary Algorithm based on I_(ϵ+)indicator(I_(ϵ+)LGEA)is developed.In I_(ϵ+)LGEA,the I_(ϵ+)indicator is incorporated into the initialization and genetic operators to guarantee the sparsity of solutions and can help improve the convergence of the algorithm.And a new constraint-handling method based on I_(ϵ+)indicator is also adopted to ensure the feasibility of solutions.The experimental results on five portfolio trading datasets including up to 1226 assets show that I_(ϵ+)LGEA outperforms some state-of-the-art MOEAs in most cases.
文摘Evolutionary computation （EC） is one of the fastest growing areas in computer science that solves intractable optimization problems by emulating biologic evolution and organizational behaviors in nature. To de- sign an EC algorithm, one needs to determine a set of algorithmic configurations like operator selections and parameter settings. How to design an effective and ef- ficient adaptation scheme for adjusting the configura- tions of EC algorithms has become a significant and promising research topic in the EC research community. This paper intends to provide a comprehensive survey on this rapidly growing field. We present a classification of adaptive EC （AEC） algorithms from the perspective of how an adaptation scheme is designed, involving the adaptation objects, adaptation evidences, and adapta- tion methods. In particular, by analyzing tile popula- tion distribution characteristics of EC algorithms, we discuss why and how the evolutionary state information of EC can be estimated and utilized for designing ef- fective EC adaptation schemes. Two AEC algorithms using the idea of evolutionary state estimation, includ- ing the clustering-based adaptive genetic algorithm and the adaptive particle swarm optimization algorithm are presented in detail. Some potential directions for the re- search of AECs are also discussed in this paper.
基金supported by the National Key Research and Development Project of China(Grant No.2018YFC1900800-5)the National Natural Science Foundation of China(Grant Nos.61773373,6153302,62021003+1 种基金61890930-5)Beijing Natural Science Foundation(Grant No.JQ19013)。
文摘Wastewater treatment plays a crucial role in alleviating water shortages and protecting the environment from pollution.Due to the strong time variabilities and complex nonlinearities within wastewater treatment systems,devising an efficient optimal controller to reduce energy consumption while ensuring effluent quality is still a bottleneck that needs to be addressed.In this paper,in order to comprehensively consider different needs of the wastewater treatment process(WTTP),a two-objective model is to consider a scope,in which minimizing energy consumption and guaranteeing effluent quality are both considered to improve wastewater treatment efficiency.To efficiently solve the model functions,a grid-based dynamic multi-objective evolutionary decomposition algorithm,namely GD-MOEA/D,is designed.A GD-MOEA/D-based intelligent optimal controller(GD-MOEA/D-IOC)is devised to achieve tracking control of the main operating variables of the WTTP.Finally,the benchmark simulation model No.1(BSM1)is applied to verify the validity of the proposed approach.The experimental results demonstrate that the constructed models can catch the dynamics of WWTP accurately.Moreover,GD-MOEA/D has better optimization ability in solving the designed models.GD-MOEA/D-IOC can achieve a significant improvement in terms of reducing energy consumption and improving effluent quality.Therefore,the designed multi-objective intelligent optimal control method for WWTP has great potential to be applied to practical engineering since it can easily achieve a highly intelligent control in WTTP.
