针对航班地面保障实际作业流程的精准刻画与性能评估,提出一种多层级航班地面保障随机Petri网(Stochastic Petri Net,SPN)构建及性能分析方法.将保障节点间的衔接过程与保障节点进行过程同时作为随机Petri网的变迁,构建多层级航班地面保...针对航班地面保障实际作业流程的精准刻画与性能评估,提出一种多层级航班地面保障随机Petri网(Stochastic Petri Net,SPN)构建及性能分析方法.将保障节点间的衔接过程与保障节点进行过程同时作为随机Petri网的变迁,构建多层级航班地面保障SPN模型解析保障全流程复杂的串并联关系.采用时间性能等价化简方法降低模型分析的难度,在此基础上建立同构马尔科夫链对模型进行性能分析,得到多层级航班地面保障流程SPN模型的库所繁忙率和变迁利用率.建立的输入输出库所繁忙率四象限图可直观揭示流程中的低效与高效运行区域,实现保障节点及衔接过程前后续状态的精准分类,结合变迁利用率结果,并基于变迁平均发生速率动态变化过程中稳态概率累计变化量识别关键保障节点或衔接,包括4个关键保障节点及6个关键节点间的衔接.研究结果表明:航班地面保障的关键部分多集中在前后衔接不紧密的不同节点间,如餐食及机供品配供完成后等待关客舱门衔接过程,即使单个保障节点的作业时间较短,但若其前续准备或后续完成状态耗时较长,易导致整体保障效率降低;当关键保障节点或衔接平均发生速率超过0.2,即对应持续时间缩短至5 min后,保障流程整体持续时间下降趋于稳定.研究成果可以为机场协同决策下目标撤轮挡时间预测准确性的提高及促进机坪运行保障效率提供理论基础和方法支撑.展开更多
Autonomous connected vehicles(ACV)involve advanced control strategies to effectively balance safety,efficiency,energy consumption,and passenger comfort.This research introduces a deep reinforcement learning(DRL)-based...Autonomous connected vehicles(ACV)involve advanced control strategies to effectively balance safety,efficiency,energy consumption,and passenger comfort.This research introduces a deep reinforcement learning(DRL)-based car-following(CF)framework employing the Deep Deterministic Policy Gradient(DDPG)algorithm,which integrates a multi-objective reward function that balances the four goals while maintaining safe policy learning.Utilizing real-world driving data from the highD dataset,the proposed model learns adaptive speed control policies suitable for dynamic traffic scenarios.The performance of the DRL-based model is evaluated against a traditional model predictive control-adaptive cruise control(MPC-ACC)controller.Results show that theDRLmodel significantly enhances safety,achieving zero collisions and a higher average time-to-collision(TTC)of 8.45 s,compared to 5.67 s for MPC and 6.12 s for human drivers.For efficiency,the model demonstrates 89.2% headway compliance and maintains speed tracking errors below 1.2 m/s in 90% of cases.In terms of energy optimization,the proposed approach reduces fuel consumption by 5.4% relative to MPC.Additionally,it enhances passenger comfort by lowering jerk values by 65%,achieving 0.12 m/s3 vs.0.34 m/s3 for human drivers.A multi-objective reward function is integrated to ensure stable policy convergence while simultaneously balancing the four key performance metrics.Moreover,the findings underscore the potential of DRL in advancing autonomous vehicle control,offering a robust and sustainable solution for safer,more efficient,and more comfortable transportation systems.展开更多
Task scheduling in cloud computing is a multi-objective optimization problem,often involving conflicting objectives such as minimizing execution time,reducing operational cost,and maximizing resource utilization.Howev...Task scheduling in cloud computing is a multi-objective optimization problem,often involving conflicting objectives such as minimizing execution time,reducing operational cost,and maximizing resource utilization.However,traditional approaches frequently rely on single-objective optimization methods which are insufficient for capturing the complexity of such problems.To address this limitation,we introduce MDMOSA(Multi-objective Dwarf Mongoose Optimization with Simulated Annealing),a hybrid that integrates multi-objective optimization for efficient task scheduling in Infrastructure-as-a-Service(IaaS)cloud environments.MDMOSA harmonizes the exploration capabilities of the biologically inspired Dwarf Mongoose Optimization(DMO)with the exploitation strengths of Simulated Annealing(SA),achieving a balanced search process.The algorithm aims to optimize task allocation by reducing makespan and financial cost while improving system resource utilization.We evaluate MDMOSA through extensive simulations using the real-world Google Cloud Jobs(GoCJ)dataset within the CloudSim environment.Comparative analysis against benchmarked algorithms such as SMOACO,MOTSGWO,and MFPAGWO reveals that MDMOSA consistently achieves superior performance in terms of scheduling efficiency,cost-effectiveness,and scalability.These results confirm the potential of MDMOSA as a robust and adaptable solution for resource scheduling in dynamic and heterogeneous cloud computing infrastructures.展开更多
Deployable Composite Thin-Walled Structures(DCTWS)are widely used in space applications due to their ability to compactly fold and self-deploy in orbit,enabled by cutouts.Cutout design is crucial for balancing structu...Deployable Composite Thin-Walled Structures(DCTWS)are widely used in space applications due to their ability to compactly fold and self-deploy in orbit,enabled by cutouts.Cutout design is crucial for balancing structural rigidity and flexibility,ensuring material integrity during large deformations,and providing adequate load-bearing capacity and stability once deployed.Most research has focused on optimizing cutout size and shape,while topology optimization offers a broader design space.However,the anisotropic properties of woven composite laminates,complex failure criteria,and multi-performance optimization needs have limited the exploration of topology optimization in this field.This work derives the sensitivities of bending stiffness,critical buckling load,and the failure index of woven composite materials with respect to element density,and formulates both single-objective and multi-objective topology optimization models using a linear weighted aggregation approach.The developed method was integrated with the commercial finite element software ABAQUS via a Python script,allowing efficient application to cutout design in various DCTWS configurations to maximize bending stiffness and critical buckling load under material failure constraints.Optimization of a classical tubular hinge resulted in improvements of 107.7%in bending stiffness and 420.5%in critical buckling load compared to level-set topology optimization results reported in the literature,validating the effectiveness of the approach.To facilitate future research and encourage the broader adoption of topology optimization techniques in DCTWS design,the source code for this work is made publicly available via a Git Hub link:https://github.com/jinhao-ok1/Topo-for-DCTWS.git.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
Understanding fish movement trajectories in aquaculture is essential for practical applications,such as disease warning,feeding optimization,and breeding management.These trajectories reveal key information about the ...Understanding fish movement trajectories in aquaculture is essential for practical applications,such as disease warning,feeding optimization,and breeding management.These trajectories reveal key information about the fish’s behavior,health,and environmental adaptability.However,when multi-object tracking(MOT)algorithms are applied to the high-density aquaculture environment,occlusion and overlapping among fish may result in missed detections,false detections,and identity switching problems,which limit the tracking accuracy.To address these issues,this paper proposes FishTracker,a MOT algorithm,by utilizing a Tracking-by-Detection framework.First,the neck part of the YOLOv8 model is enhanced by introducing a Multi-Scale Dilated Attention(MSDA)module to improve object localization and classification confidence.Second,an Adaptive Kalman Filter(AKF)is employed in the tracking phase to dynamically adjust motion prediction parameters,thereby overcoming target adhesion and nonlinear motion in complex scenarios.Experimental results show that FishTracker achieves a multi-object tracking accuracy(MOTA)of 93.22% and 87.24% in bright and dark illumination conditions,respectively.Further validation in a real aquaculture scenario reveal that FishTracker achieves aMOTA of 76.70%,which is 5.34% higher than the baselinemodel.The higher order tracking accuracy(HOTA)reaches 50.5%,which is 3.4% higher than the benchmark.In conclusion,FishTracker can provide reliable technical support for accurate tracking and behavioral analysis of high-density fish populations.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
Federated learning is a distributed framework that trains a centralised model using data from multiple clients without transferring that data to a central server.Despite rapid progress,federated learning still faces s...Federated learning is a distributed framework that trains a centralised model using data from multiple clients without transferring that data to a central server.Despite rapid progress,federated learning still faces several unsolved challenges.Specifically,communication costs and system heterogeneity,such as nonidentical data distribution,hinder federated learning's progress.Several approaches have recently emerged for federated learning involving heterogeneous clients with varying computational capabilities(namely,heterogeneous federated learning).However,heterogeneous federated learning faces two key challenges:optimising model size and determining client selection ratios.Moreover,efficiently aggregating local models from clients with diverse capabilities is crucial for addressing system heterogeneity and communication efficiency.This paper proposes an evolutionary multiobjective optimisation framework for heterogeneous federated learning(MOHFL)to address these issues.Our approach elegantly formulates and solves a biobjective optimisation problem that minimises communication cost and model error rate.The decision variables in this framework comprise model sizes and client selection ratios for each Q client cluster,yielding a total of 2×Q optimisation parameters to be tuned.We develop a partition-based strategy for MOHFL that segregates clients into clusters based on their communication and computation capabilities.Additionally,we implement an adaptive model sizing mechanism that dynamically assigns appropriate subnetwork architectures to clients based on their computational constraints.We also propose a unified aggregation framework to combine models of varying sizes from heterogeneous clients effectively.Extensive experiments on multiple datasets demonstrate the effectiveness and superiority of our proposed method compared to existing approaches.展开更多
Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrain...Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrained onboard devices to nearby Roadside Unit(RSU),thereby achieving lower delay and energy consumption.However,due to the limited storage capacity and energy budget of RSUs,it is challenging to meet the demands of the highly dynamic Internet of Vehicles(IoV)environment.Therefore,determining reasonable service caching and computation offloading strategies is crucial.To address this,this paper proposes a joint service caching scheme for cloud-edge collaborative IoV computation offloading.By modeling the dynamic optimization problem using Markov Decision Processes(MDP),the scheme jointly optimizes task delay,energy consumption,load balancing,and privacy entropy to achieve better quality of service.Additionally,a dynamic adaptive multi-objective deep reinforcement learning algorithm is proposed.Each Double Deep Q-Network(DDQN)agent obtains rewards for different objectives based on distinct reward functions and dynamically updates the objective weights by learning the value changes between objectives using Radial Basis Function Networks(RBFN),thereby efficiently approximating the Pareto-optimal decisions for multiple objectives.Extensive experiments demonstrate that the proposed algorithm can better coordinate the three-tier computing resources of cloud,edge,and vehicles.Compared to existing algorithms,the proposed method reduces task delay and energy consumption by 10.64%and 5.1%,respectively.展开更多
文摘针对航班地面保障实际作业流程的精准刻画与性能评估,提出一种多层级航班地面保障随机Petri网(Stochastic Petri Net,SPN)构建及性能分析方法.将保障节点间的衔接过程与保障节点进行过程同时作为随机Petri网的变迁,构建多层级航班地面保障SPN模型解析保障全流程复杂的串并联关系.采用时间性能等价化简方法降低模型分析的难度,在此基础上建立同构马尔科夫链对模型进行性能分析,得到多层级航班地面保障流程SPN模型的库所繁忙率和变迁利用率.建立的输入输出库所繁忙率四象限图可直观揭示流程中的低效与高效运行区域,实现保障节点及衔接过程前后续状态的精准分类,结合变迁利用率结果,并基于变迁平均发生速率动态变化过程中稳态概率累计变化量识别关键保障节点或衔接,包括4个关键保障节点及6个关键节点间的衔接.研究结果表明:航班地面保障的关键部分多集中在前后衔接不紧密的不同节点间,如餐食及机供品配供完成后等待关客舱门衔接过程,即使单个保障节点的作业时间较短,但若其前续准备或后续完成状态耗时较长,易导致整体保障效率降低;当关键保障节点或衔接平均发生速率超过0.2,即对应持续时间缩短至5 min后,保障流程整体持续时间下降趋于稳定.研究成果可以为机场协同决策下目标撤轮挡时间预测准确性的提高及促进机坪运行保障效率提供理论基础和方法支撑.
基金the Hebei Province Science and Technology Plan Project(19221909D)rincess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R308),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Autonomous connected vehicles(ACV)involve advanced control strategies to effectively balance safety,efficiency,energy consumption,and passenger comfort.This research introduces a deep reinforcement learning(DRL)-based car-following(CF)framework employing the Deep Deterministic Policy Gradient(DDPG)algorithm,which integrates a multi-objective reward function that balances the four goals while maintaining safe policy learning.Utilizing real-world driving data from the highD dataset,the proposed model learns adaptive speed control policies suitable for dynamic traffic scenarios.The performance of the DRL-based model is evaluated against a traditional model predictive control-adaptive cruise control(MPC-ACC)controller.Results show that theDRLmodel significantly enhances safety,achieving zero collisions and a higher average time-to-collision(TTC)of 8.45 s,compared to 5.67 s for MPC and 6.12 s for human drivers.For efficiency,the model demonstrates 89.2% headway compliance and maintains speed tracking errors below 1.2 m/s in 90% of cases.In terms of energy optimization,the proposed approach reduces fuel consumption by 5.4% relative to MPC.Additionally,it enhances passenger comfort by lowering jerk values by 65%,achieving 0.12 m/s3 vs.0.34 m/s3 for human drivers.A multi-objective reward function is integrated to ensure stable policy convergence while simultaneously balancing the four key performance metrics.Moreover,the findings underscore the potential of DRL in advancing autonomous vehicle control,offering a robust and sustainable solution for safer,more efficient,and more comfortable transportation systems.
文摘Task scheduling in cloud computing is a multi-objective optimization problem,often involving conflicting objectives such as minimizing execution time,reducing operational cost,and maximizing resource utilization.However,traditional approaches frequently rely on single-objective optimization methods which are insufficient for capturing the complexity of such problems.To address this limitation,we introduce MDMOSA(Multi-objective Dwarf Mongoose Optimization with Simulated Annealing),a hybrid that integrates multi-objective optimization for efficient task scheduling in Infrastructure-as-a-Service(IaaS)cloud environments.MDMOSA harmonizes the exploration capabilities of the biologically inspired Dwarf Mongoose Optimization(DMO)with the exploitation strengths of Simulated Annealing(SA),achieving a balanced search process.The algorithm aims to optimize task allocation by reducing makespan and financial cost while improving system resource utilization.We evaluate MDMOSA through extensive simulations using the real-world Google Cloud Jobs(GoCJ)dataset within the CloudSim environment.Comparative analysis against benchmarked algorithms such as SMOACO,MOTSGWO,and MFPAGWO reveals that MDMOSA consistently achieves superior performance in terms of scheduling efficiency,cost-effectiveness,and scalability.These results confirm the potential of MDMOSA as a robust and adaptable solution for resource scheduling in dynamic and heterogeneous cloud computing infrastructures.
基金supported by the National Natural Science Foundation of China(No.12202295)the International(Regional)Cooperation and Exchange Projects of the National Natural Science Foundation of China(No.W2421002)+2 种基金the Sichuan Science and Technology Program(No.2025ZNSFSC0845)Zhejiang Provincial Natural Science Foundation of China(No.ZCLZ24A0201)the Fundamental Research Funds for the Provincial Universities of Zhejiang(No.GK249909299001-004)。
文摘Deployable Composite Thin-Walled Structures(DCTWS)are widely used in space applications due to their ability to compactly fold and self-deploy in orbit,enabled by cutouts.Cutout design is crucial for balancing structural rigidity and flexibility,ensuring material integrity during large deformations,and providing adequate load-bearing capacity and stability once deployed.Most research has focused on optimizing cutout size and shape,while topology optimization offers a broader design space.However,the anisotropic properties of woven composite laminates,complex failure criteria,and multi-performance optimization needs have limited the exploration of topology optimization in this field.This work derives the sensitivities of bending stiffness,critical buckling load,and the failure index of woven composite materials with respect to element density,and formulates both single-objective and multi-objective topology optimization models using a linear weighted aggregation approach.The developed method was integrated with the commercial finite element software ABAQUS via a Python script,allowing efficient application to cutout design in various DCTWS configurations to maximize bending stiffness and critical buckling load under material failure constraints.Optimization of a classical tubular hinge resulted in improvements of 107.7%in bending stiffness and 420.5%in critical buckling load compared to level-set topology optimization results reported in the literature,validating the effectiveness of the approach.To facilitate future research and encourage the broader adoption of topology optimization techniques in DCTWS design,the source code for this work is made publicly available via a Git Hub link:https://github.com/jinhao-ok1/Topo-for-DCTWS.git.
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.
基金funded by the Fundamental Research Funds for the Central Universities(Grant No.106-YDZX2025022)the Startup Foundation of New Professor at Nanjing Agricultural University(Grant No.106-804005)the“Qing Lan Project”of Jiangsu Higher Education Institutions.
文摘Understanding fish movement trajectories in aquaculture is essential for practical applications,such as disease warning,feeding optimization,and breeding management.These trajectories reveal key information about the fish’s behavior,health,and environmental adaptability.However,when multi-object tracking(MOT)algorithms are applied to the high-density aquaculture environment,occlusion and overlapping among fish may result in missed detections,false detections,and identity switching problems,which limit the tracking accuracy.To address these issues,this paper proposes FishTracker,a MOT algorithm,by utilizing a Tracking-by-Detection framework.First,the neck part of the YOLOv8 model is enhanced by introducing a Multi-Scale Dilated Attention(MSDA)module to improve object localization and classification confidence.Second,an Adaptive Kalman Filter(AKF)is employed in the tracking phase to dynamically adjust motion prediction parameters,thereby overcoming target adhesion and nonlinear motion in complex scenarios.Experimental results show that FishTracker achieves a multi-object tracking accuracy(MOTA)of 93.22% and 87.24% in bright and dark illumination conditions,respectively.Further validation in a real aquaculture scenario reveal that FishTracker achieves aMOTA of 76.70%,which is 5.34% higher than the baselinemodel.The higher order tracking accuracy(HOTA)reaches 50.5%,which is 3.4% higher than the benchmark.In conclusion,FishTracker can provide reliable technical support for accurate tracking and behavioral analysis of high-density fish populations.
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 by the National Research Foundation of Korea grant funded by the Korea government(RS-2023-00217116)。
文摘Federated learning is a distributed framework that trains a centralised model using data from multiple clients without transferring that data to a central server.Despite rapid progress,federated learning still faces several unsolved challenges.Specifically,communication costs and system heterogeneity,such as nonidentical data distribution,hinder federated learning's progress.Several approaches have recently emerged for federated learning involving heterogeneous clients with varying computational capabilities(namely,heterogeneous federated learning).However,heterogeneous federated learning faces two key challenges:optimising model size and determining client selection ratios.Moreover,efficiently aggregating local models from clients with diverse capabilities is crucial for addressing system heterogeneity and communication efficiency.This paper proposes an evolutionary multiobjective optimisation framework for heterogeneous federated learning(MOHFL)to address these issues.Our approach elegantly formulates and solves a biobjective optimisation problem that minimises communication cost and model error rate.The decision variables in this framework comprise model sizes and client selection ratios for each Q client cluster,yielding a total of 2×Q optimisation parameters to be tuned.We develop a partition-based strategy for MOHFL that segregates clients into clusters based on their communication and computation capabilities.Additionally,we implement an adaptive model sizing mechanism that dynamically assigns appropriate subnetwork architectures to clients based on their computational constraints.We also propose a unified aggregation framework to combine models of varying sizes from heterogeneous clients effectively.Extensive experiments on multiple datasets demonstrate the effectiveness and superiority of our proposed method compared to existing approaches.
基金supported by Key Science and Technology Program of Henan Province,China(Grant Nos.242102210147,242102210027)Fujian Province Young and Middle aged Teacher Education Research Project(Science and Technology Category)(No.JZ240101)(Corresponding author:Dong Yuan).
文摘Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrained onboard devices to nearby Roadside Unit(RSU),thereby achieving lower delay and energy consumption.However,due to the limited storage capacity and energy budget of RSUs,it is challenging to meet the demands of the highly dynamic Internet of Vehicles(IoV)environment.Therefore,determining reasonable service caching and computation offloading strategies is crucial.To address this,this paper proposes a joint service caching scheme for cloud-edge collaborative IoV computation offloading.By modeling the dynamic optimization problem using Markov Decision Processes(MDP),the scheme jointly optimizes task delay,energy consumption,load balancing,and privacy entropy to achieve better quality of service.Additionally,a dynamic adaptive multi-objective deep reinforcement learning algorithm is proposed.Each Double Deep Q-Network(DDQN)agent obtains rewards for different objectives based on distinct reward functions and dynamically updates the objective weights by learning the value changes between objectives using Radial Basis Function Networks(RBFN),thereby efficiently approximating the Pareto-optimal decisions for multiple objectives.Extensive experiments demonstrate that the proposed algorithm can better coordinate the three-tier computing resources of cloud,edge,and vehicles.Compared to existing algorithms,the proposed method reduces task delay and energy consumption by 10.64%and 5.1%,respectively.
