大模型与智能优化算法集成研究综述
CSTR:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

N945

基金项目:

陕西省科技创新团队项目(2023-CX-TD-07);陕西省重点研发计划项目(2024GH-ZDXM-48);广东省重点领域研发计划项目(2021ZDZX1019);航空科学基金项目(2023Z034053004);国家自然科学基金项目(62203365, 62203461);中国博士后科学基金项目(2023M742843);陕西省科协青年人才托举计划项目(20220121, 20230125);智能化航天测运控教育部重点实验室基金项目(CYK2024-02-04).


A research review on integration of large models and intelligent optimization algorithms
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    大模型作为人工智能领域的一项突破性进展, 在新一轮全球科技革命和产业变革中发挥着重要作用. 智能优化算法凭借其在降本增效方面的优势, 极大地推动了社会经济的稳步发展, 两者的有机结合有望为应对复杂交叉的科学研究和工程实践注入新鲜血液. 鉴于此, 提出大模型与智能优化算法间相互赋能的综述. 首先, 从定义和分类两个方面介绍大模型和智能优化算法; 然后, 从大模型赋能智能优化算法和智能优化算法赋能大模型两条路线梳理最新研究进展: 前者围绕代理辅助优化、自动优化建模、自动算法设计与生成、自动算法分析与改进、行业应用开展, 后者基于参数优化、提示优化、联合优化进行, 从通用基础和专用应用两个视角擘画两者的发展方向; 最后, 展望大模型与智能优化算法集成的机遇和挑战.

    Abstract:

    As a breakthrough in the field of artificial intelligence, large models play a crucial role in the new round of global technological revolution and industrial transformation. Intelligent optimization algorithms have greatly promoted the steady development of social economy by virtue of their advantages in reducing cost and upgrading efficiency. The integration of both is expected to inject fresh blood into handling scientific research and engineering practice with the trait of being complex and multidisciplinary. This article reviews the empowerment between foundational models and intelligent optimization algorithms. First, large models and intelligent optimization algorithms are profiled from two aspects: definition and classification. Then, the latest research progress is combed from two routes: one is empowering large models with intelligent optimization algorithms, the other is empowering intelligent optimization algorithms with large models. Wherein, the former revolves around surrogate-assisted optimization, automatic optimization modeling, automatic algorithm designing and generating, automatic algorithm analyzing and enhancing, and industry practices, while the latter's report is according to parameter optimization, prompt optimization, and joint optimization, the developing trend for both of which is envisioned from the perspectives of general basics and specific applications. The article concludes with prospecting the integration of foundation models and intelligent optimization algorithms.

    参考文献
    相似文献
    引证文献
引用本文

张浩然,李君,邢立宁,等.大模型与智能优化算法集成研究综述[J].控制与决策,2026,41(4):871-891

复制
相关视频

分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2025-01-24
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2026-03-24
  • 出版日期: 2026-04-10
文章二维码