报告题目：Network-Aware Peer-to-Peer Cyber Risk Sharing with Risk Propagation

报告所属学科：管理科学与工程

报告人：徐茂超（Illinois State University）

报告时间：2026年6月22日 14:00-16:00

报告地点：经管学院702室

报告摘要：

In this talk, we discuss a network-aware framework for peer-to-peer (P2P) cyber risk sharing under risk propagation. The framework distinguishes between an attack network, which governs compromise contagion and cross-firm loss dependence, and a sharing network, which determines how residual cyber burdens are allocated across participants. The proposed sharing rule decomposes each claimant's loss into three components: a causal-attribution layer, a self-retention layer, and a residual-pooling layer. We formulate the design of the residual-pooling rule as a weighted burden-variance minimization problem subject to full allocation and actuarial fairness. We derive a closed-form benchmark and provide a graph-based feasibility condition for the existence of a nonnegative fair pooling rule. When loss moments are not directly available, we induce dependence from a multi-hop cyber propagation model and derive exact compromise probabilities, an explicit attack-based attribution matrix, and closed-form covariance expressions for special network structures. Numerical examples and a supply-chain case study show that the optimized P2P rule can substantially stabilize realized burdens while preserving actuarial fairness. These results suggest that network-aware P2P design can provide a tractable and interpretable mechanism for complementing centralized cyber insurance by aligning cyber risk sharing with propagation-driven dependence.

报告人简介：

徐茂超教授是国际网络风险量化与保险精算交叉领域的知名学者，主要研究方向为网络安全风险量化、保险科技与数字风险治理。现任美国伊利诺伊州立大学数学系终身教授、OSF HealthCare 医疗集团与伊利诺伊州立大学数据科学方向带头人，2010 年获美国波特兰州立大学统计学博士学位。徐茂超教授先后主持美国精算师协会（SOA）、北美产险精算学会（CAS）等行业权威机构资助的十余项科研项目，在《Annals of Applied Statistics》《Technometrics》《IEEE TIFS》《IME》《EJOR》等应用统计、信息安全、保险精算领域顶级期刊发表论文 70 余篇。成果曾获北美精算师协会最佳论文奖、伊利诺伊州立大学杰出学术成就奖等重要荣誉，相关研究获北美产险精算学会官方期刊专题专访报道。徐茂超教授现任美国 Rankiteo 风险管理公司机器学习部门负责人，牵头搭建企业级网络风险智能评估与核保平台，覆盖百万级市场主体风险量化；曾兼任 Safe Security、InsurAce、CloudCover 等多家国际头部网络安全与保险企业学术顾问，为产品设计、定价模型与风险治理提供核心技术支撑。