AI For Science 超大规模智能优化求解器
2024-12-27 04:16
清华大学计算机系智能优化研究团队(https://thuiar.github.io/)依托北京信息科学与技术国家研究中心等科研平台。团队负责人以第一或通讯作者发表CCF A类国际会议论文25篇;SCI国际期刊论文45篇(一区32篇,IF>5.0有39篇,ESI高被引2篇),SCI 引用2425次,谷歌学术引用6130次;编著教材4部,学术专著4部;获发明专利授权36项,软著授权26项。成果已被ACM Comput. Surv.等顶级期刊引用,引用者包括6位海内外科学院院士和30余位IEEE Fellow。受邀担任工信部03号科技重大专项核心专家、“科技助力经济2020”国家重点研发计划项目验收专家组组长、国家发改委国家产业创新中心咨询专家、Elsevier人工智能国际期刊Intell. Syst. Appl.主编和Expert Syst. Appl.(1区,SCI IF=8.665)副主编。曾获得国家科技进步二等奖1次,北京市科学技术一等奖1次,行业协会科技一等奖2次,其它省部级二、三等奖3次。
团队近年来专注于智能优化理论和作业调度方法的研究工作,提出了面向多目标优化的智能演化优化理论和柔性作业车间调度(FJS)方法技术。所提出的基于Theta支配关系的演化优化方法被IEEE智能系统学会副主席评价为自2010年继第三代非支配排序算法(NSGA-III)之后又一里程碑性的最先进研究成果,有效解决了高维多目标优化问题求解,是当前最先进方法之一;成果发表在IEEE Trans. on EC(SCI IF= 16.497)等期刊上,被美国阿贡国家实验室、橡树岭国家实验室、日本宇航局和普林斯顿大学等应用于专业问题的优化求解。另一方面,所提出的作业调度方法成果被欧洲科学院院士焦李成教授评价为当前五类经典FJS调度方法之一,国际权威分析报告LCR分析为当前十大经典FJS调度算法排名第4。成功应用于国家面临的“卡脖子”领域——芯片制造领域自主国产化装备上(首台国产高密度等离子刻蚀机、PVD等),曾获国家科技进步二等奖和省部级和行业协会科技奖。独立编著出版国内演化机器学习领域首部学术专著《演化机器学习》。即将出版“演化学习与智能优化”系列学术专著。
Computer science department, TsingHua university (https://thuiar.github.io/) based on intelligent optimization research team in Beijing information science and technology, national research center for scientific research platform. The team leader published 25 CCF Class A international conference papers as the first or corresponding author; 45 SCI international journal papers (32 in Region I, 39 IF>5.0, 2 ESI highly cited), SCI citations, Google academic citations 6130 times; He has edited 4 textbooks and 4 academic monographs. It has been authorized 36 invention patents and 26 soft patents. The results have been cited by top journals such as ACM Comput. Surv., including 6 academicians of Chinese Academy of Sciences at home and abroad and more than 30 IEEE Fellows. Invited to serve as the core expert of No. 03 Science and Technology major project of the Ministry of Industry and Information Technology, "Science and Technology to help the economy 2020" national key research and development plan project acceptance team leader, National Industrial Innovation Center of the National Development and Reform Commission consulting expert, Elsevier artificial intelligence international journal Intell.Syst.Appl. Chief Editor and Associate Editor of Expert Syst.Appl. (Region 1, SCI IF=8.665). He has won the second prize of National Science and Technology Progress once, the first prize of Beijing Science and Technology once, the first prize of industry association science and technology twice, and the second and third prizes of other provincial and ministerial levels three times.
In recent years, the team has focused on the research of intelligent optimization theory and job scheduling methods, and proposed intelligent evolutionary optimization theory for multi-objective optimization and flexible job shop scheduling (FJS) method technology. The proposed evolutionary optimization method based on Theta domination relationship was evaluated by the vice president of IEEE Intelligent Systems Society as another milestone of the most advanced research achievement since 2010 after the third generation of non-domination sorting algorithm (NSGA-III), which effectively solved the high-dimensional multi-objective optimization problem, and is one of the most advanced methods at present. The results have been published in IEEE Trans.on EC (SCI IF= 16.497) and other journals, and have been applied to the optimization of professional problems by Argonne National Laboratory, Oak Ridge National Laboratory, Japan Space Agency and Princeton University. On the other hand, the proposed job scheduling method was evaluated as one of the five classic FJS scheduling methods by Professor Jiao Licheng, academician of the European Academy of Sciences, and the LCR analysis of the international authoritative analysis report ranked the fourth among the top ten classic FJS scheduling algorithms. It has been successfully applied to the "jam neck" field faced by the country - the independent localization equipment in the field of chip manufacturing (the first domestic high-density plasma etching machine, PVD, etc.), and has won the second prize of National science and Technology Progress and the science and technology award of provincial and ministerial level and industry association. He independently edited and published the first academic monograph "Evolutionary Machine Learning" in the field of evolutionary machine learning in China. A series of academic monographs "Evolutionary Learning and intelligent Optimization" will be published soon.