报告题目：面向分布式智能的博弈学习理论与方法（Game-theoretical Learning in the Distributed Intelligence Era）
报告人姓名：谭少林
报告人单位：中关村实验室
报告摘要：Evolutionary games and dynamics are finding increasingly applications in design learning and control protocols for a variety of multi-agent problems. The implicitly requirement for full communication has been a main limitation of the evolutionary game dynamic approach in engineering tasks with various information constraints. In this lecture, we will report several methods to build distributed game-learning dynamics with both static or dynamical graphical communication structures. The idea of local competition and game decoupling will be strengthened, and the convergence results of the obtained distributed game-learning dynamics will be given. Some typical multi-agent problems are also given for illustration and several future problems in this area are listed in the end.
个人简历：谭少林，2014年博士毕业于中国科学院数学与系统科学研究院；之后进入湖南大学电气与信息工程学院工作，先后任助理教授（2014-2016）、副教授/博士生导师（2017-2022）；2022年起进入北京中关村实验室工作。出版专著《复杂网络上的博弈及其演化动力学》一部，在IEEE TAC/TNNLS/ /TKDE等期刊上发表论文50余篇。获国家优秀青年科学基金、湖南省优秀青年科学基金、中国指挥与控制学会青年科学家奖、中科院数学与系统科学研究院突出科研成果奖等荣誉。主要研究领域包括：分布式博弈与决策、工业互联网人机物交互、图数据表示与学习等。