面向复杂帕累托前沿的自适应策略与算法研究
结题报告
批准号:
62006074
项目类别:
青年科学基金项目
资助金额:
24.0 万元
负责人:
王鹏
依托单位:
学科分类:
智能系统与人工智能安全
结题年份:
2023
批准年份:
2020
项目状态:
已结题
项目参与者:
王鹏
国基评审专家1V1指导 中标率高出同行96.8%
结合最新热点,提供专业选题建议
深度指导申报书撰写,确保创新可行
指导项目中标800+,快速提高中标率
客服二维码
微信扫码咨询
中文摘要
具有复杂帕累托前沿的多目标优化是科学研究与工程实践中普遍存在的一类优化问题,其求解算法研究难度大、应用前景广,目前缺少成熟的应对策略与方法,具有重要的理论研究意义与应用价值。本项目将子问题权重向量调整与计算资源分配策略相结合,从自适应策略的机制出发研究解决复杂帕累托前沿多目标优化问题的新方法,主要研究自适应资源分配策略的框架与理论特性、迭代策略与算法以及并行实现方法,其关键思路是利用子问题与对应解的关系以及子问题更新迭代的特点,针对多目标优化问题的复杂帕累托前沿形状特征,设计自适应策略与机制引导进化种群主动应对不规则变化的帕累托前沿形状。项目采用理论分析与算法实验设计相结合的研究方法,目标是为复杂帕累托前沿多目标优化问题探索新的求解思路与方法,以推动该领域的研究与发展。
英文摘要
Multi-objective optimization with complex Pareto frontiers is a type of optimization problems commonly found in scientific research and engineering practice. Its solution algorithms are difficult to study and it have broad application prospects. there are currently no mature coping strategies and methods. It has important theoretical research significance and application value. This project combines the sub-problem weight vector adjustment with the calculation resource allocation strategy, and starts from the mechanism of the adaptive strategy to study a new method to solve the complex Pareto frontier multi-objective optimization problem. It mainly studies the framework and theoretical characteristics of the adaptive resource allocation strategy, iterative strategies and algorithms, and parallel implementation methods. The key idea is to design adaptive strategies and mechanisms to guide the evolutionary population to actively deal with the irregular shape of the pareto frontier by making use of the relationship between the subproblem and the corresponding solution and the characteristics of the update iteration of the subproblem, in view of complex Pareto front shape characteristics of multi-objective optimization problems. The project adopts a research method combining theoretical analysis and algorithm experimental design. The goal is to explore new solutions and methods for complex Pareto frontier multi-objective optimization problems in order to promote research and development in this field.
期刊论文列表
专著列表
科研奖励列表
会议论文列表
专利列表
DOI:10.1093/bib/bbab319
发表时间:2021-08-10
期刊:BRIEFINGS IN BIOINFORMATICS
影响因子:9.5
作者:Cai, Lijun;Lu, Changcheng;Su, Yansen
通讯作者:Su, Yansen
Data augmentation based semi-supervised method to improve COVID-19 CT classification.
基于数据增强的半监督方法改进 COVID-19 CT 分类。
DOI:10.3934/mbe.2023294
发表时间:2023
期刊:Mathematical biosciences and engineering : MBE
影响因子:--
作者:Xiangtao Chen;Y. Bai;Peng Wang;Jiawei Luo
通讯作者:Jiawei Luo
DOI:10.1038/s41540-023-00267-8
发表时间:2023-02-10
期刊:NPJ systems biology and applications
影响因子:4
作者:
通讯作者:
DOI:10.1109/tcbb.2022.3203185
发表时间:2023-03-01
期刊:IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
影响因子:4.5
作者:Li, Xiong;Lin, Yangkai;Zhou, Juan
通讯作者:Zhou, Juan
DOI:10.1016/j.asoc.2022.109392
发表时间:2022-07
期刊:Appl. Soft Comput.
影响因子:--
作者:Bing-chuan Wang;Zhi-Zhong Liu;Wu Song
通讯作者:Bing-chuan Wang;Zhi-Zhong Liu;Wu Song
基于生成对抗网络的多目标优化算法研究
  • 批准号:
    2021JJ40120
  • 项目类别:
    省市级项目
  • 资助金额:
    0.0万元
  • 批准年份:
    2021
  • 负责人:
    王鹏
  • 依托单位:
国内基金
海外基金