典型文献
A Dynamic Resource Allocation Strategy with Reinforcement Learning for Multimodal Multi-objective Optimization
文献摘要：
Many isolation approaches,such as zoning search,have been proposed to preserve the diversity in the decision space of mul-timodal multi-objective optimization(MMO).However,these approaches allocate the same computing resources for subspaces with dif-ferent difficulties and evolution states.In order to solve this issue,this paper proposes a dynamic resource allocation strategy(DRAS)with reinforcement learning for multimodal multi-objective optimization problems(MMOPs).In DRAS,relative contribution and im-provement are utilized to define the aptitude of subspaces,which can capture the potentials of subspaces accurately.Moreover,the rein-forcement learning method is used to dynamically allocate computing resources for each subspace.In addition,the proposed DRAS is applied to zoning searches.Experimental results demonstrate that DRAS can effectively assist zoning search in finding more and better distributed equivalent Pareto optimal solutions in the decision space.
文献关键词：
中图分类号：
作者姓名：
Qian-Long Dang;Wei Xu;Yang-Fei Yuan
作者机构：
School of Mathematics and Statistics,Xidian University,Xi'an 710126,China
文献出处：
引用格式：
[1]Qian-Long Dang;Wei Xu;Yang-Fei Yuan-.A Dynamic Resource Allocation Strategy with Reinforcement Learning for Multimodal Multi-objective Optimization)[J].机器智能研究（英文）,2022(02):138-152
A类：
timodal,DRAS,MMOPs
B类：
Dynamic,Resource,Allocation,Strategy,Reinforcement,Learning,Multimodal,objective,Optimization,Many,isolation,approaches,such,zoning,have,been,proposed,preserve,diversity,decision,optimization,However,these,allocate,same,computing,resources,subspaces,ferent,difficulties,evolution,states,In,order,solve,this,issue,paper,proposes,allocation,strategy,reinforcement,learning,multimodal,problems,relative,contribution,provement,utilized,define,aptitude,which,can,capture,potentials,accurately,Moreover,method,used,dynamically,each,addition,applied,searches,Experimental,results,demonstrate,that,effectively,assist,finding,more,better,distributed,equivalent,Pareto,optimal,solutions
AB值：
0.545731
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
