[PDF][PDF] 基于改进演化博弈模型的网络防御决策方法
马润年， 张恩宁， 王刚， 马宇峰， 翁江 - 电子与信息学报, 2023 - jeit.ac.cn
… hypothesis to accelerate the convergence of the model and … , and the improved replication
dynamics equation is proposed … can converge to the -neighborhood of the Nash equilibrium …
dynamics equation is proposed … can converge to the -neighborhood of the Nash equilibrium …
[PDF][PDF] 基于微分对策理论的非线性控制回顾与展望
谭拂晓， 刘德荣， 关新平， 罗斌 - 自动化学报, 2014 - aas.net.cn
… Differential game is a mathematical tool for dealing with the problems of continuous
dynamic conflict, competition or cooperation with two or more control actions using differential …
dynamic conflict, competition or cooperation with two or more control actions using differential …
面向多智能体博弈对抗的对手建模框架
罗俊仁， 张万鹏， 袁唯淋， 胡振震， 陈少飞… - 系统仿真学报, 2022 - china-simulation.com
… 在线无悔学习方法,其将在线序 贯决策过程构建成在线/对抗马尔可夫决策过程 (online/adversarial
MDP)模型[64-65],通过采样生成 多个策略,采用多臂机在线凸优化控制动态后悔值 (dynamic …
MDP)模型[64-65],通过采样生成 多个策略,采用多臂机在线凸优化控制动态后悔值 (dynamic …