报告题目:Evolutionary Learning: From Theory to Practice
报告时间:2024年11月4日14:30
报告地点:威尼斯432888camE202
报告人:钱超
报告人单位:南京大学
报告人简介:钱超,南京大学人工智能学院教授、博导。长期从事人工智能中演化学习基础理论研究,以第一/通讯作者在人工智能国际一流期刊和会议上发表50余篇论文,出版专著《Evolutionary Learning》,获ACM GECCO’11最佳理论论文奖,受邀担任IEEE计算智能学会“演化算法理论分析”工作组主席,获CCF-IEEE CS青年科学家奖(2023)。部分成果成功应用于华为工厂排产、无线网络优化、芯片寄存器寻优等任务,获2次华为“难题揭榜”火花奖,落地华为产品线;应用于自然科学基础问题(如土壤微生物源碳预测),成果以共同一作发表于美国国家科学院院刊PNAS。担任人工智能/演化计算权威国际期刊Artificial Intelligence、Evolutionary Computation、IEEE Trans. Evolutionary Computation等编委,在国际人工智能联合大会IJCAI’22作Early Career Spotlight报告,并将担任第22届亚太人工智能国际会议PRICAI’25程序委员会主席。获国家优秀青年科学基金(2020),并主持新一代人工智能国家科技重大专项(青年科学家)。指导本科生获国家自然本科生项目,执教《启发式搜索与演化算法》被研究生选为“我心目中的好课程”,获南京大学青年五四奖章、“师德先进”青年教师奖。
报告摘要:Machine learning tasks often involve complex optimization, like black-box and multi-objective optimization, which may make conventional optimization algorithms fail. Evolutionary algorithms, inspired by Darwin’s theory of evolution, have yielded encouraging outcomes. However, due to their heuristic nature, most outcomes to date have been empirical and lack theoretical support. In this talk, I will introduce our efforts towards building the theoretical foundation of evolutionary learning and developing better algorithms inspired by theories. Finally, I will introduce some successful applications in industry (e.g., electronic design automation) and science (e.g., studying the origin and evolution of life).