报 告 人： 李泉林 教授
报告题目：Markov Processes in Blockchain Systems
报告时间：2020年1月8日（周三）下午4：30
报告地点：静远楼1506报告厅
主办单位：数学与统计学院、科学技术研究院
报告人简介：
李泉林，北京工业大学经管学院教授、博士生导师，研究领域包括随机模型、随机过程、博弈论、排队论、计算机网络、网络安全、网络资源管理、网络信息理论、超市模型、负载调配模型、RFID技术、物联网、大数据、云计算、区块链、数据中心、医疗服务系统、共享经济、制造系统、供应链管理等方面。他在重要的国际学术刊物上发表了60余篇SCI学术论文；在Springer出版专著《Constructive Computation in Stochastic Models with Applications:RG-Factorizations》；在Springer主编论文集《Editorial for the special issue: Retrial Queues (WRQ'2010), Guest Editors，Operational Research: An International Journal，2012》、《Queueing Theory and Network Applications. Lecture Notes in Computer Science, Volume 10591，2017》、《Stochastic Models in Reliability,Network Security and System Safety: Dedicated to Jinhua Cao on the Occasion of His 80th Birthday. Communications in Computer and Information Science, Volume1102, 2019》；30余次担任排队论、随机模型与应用概率等领域重要国际学术会议的学术委员会委员（4次大会主席）；获得了2004年教育部新世纪优秀人才、2005年教育部自然科学一等奖、2007年北京市科学技术二等奖、2008年北京市精品课、2013年河北省科技领军人才计划、2014年河北省科学技术二等奖、2015年INFORMS优秀论文奖、2018年第7届计算社会网络国际会议（CSoNet2018）唯一最佳论文奖。
报告摘要：
This talk focuses on our recent research on Markov Processes in Blockchain Systems. The blockchain systems are established as multi-dimensional Markov processes by means of the longest chain rule of chain-fork structure. We address several interesting issues or topics related to the multi-dimensional Markov processes. This further sets up mathematical models and develops economic theory of blockchain. Here, we shall care for:
(1) How to study the multi-dimensional Markov processes, for example, stable conditions, steady-state probability, first passage time, sojourn time and so forth. Perhaps the Markov processes bring you to enter a queer theoretical space from such an interesting practical technology.
(2) Block reward, transaction fee and their allocation methods greatly motivate many miners in a blockchain to take shape some selfish mining alliances evolutionarily, while the selfish mining alliances will lead to various attacks on security of blockchain. As a first exploration, we provide a unified and comprehensive framework for expressing the attacks grown out of the selfish mining alliances, a physical structure of which is given a detailed observation and interpretation in terms of the Markov processes. This may be viewed as a key improvement in the study of blockchain mining management. On the other hand, our method can also be developed to analyze blochchain systems through some simple and intuitive applications of Markov decision processes and stochastic game modeling.
(3) We show that the multi-dimensional Markov processes will play an important role in the study of blockchain systems and in the design of consensus mechanism of related distributed systems. Also, they can motivate a series of promising future research on development of blockchain technologies.
