Learning of Dynamical Systems under Adversarial Attacks - Null Space Property Perspective

Learning of Dynamical Systems under Adversarial Attacks - Null Space Property Perspective
复制标题

对抗性攻击下动态系统的学习 - 零空间属性视角

DOI:
--
复制
发表时间:
2022
期刊:
American Control Conference
影响因子:
--
通讯作者:
J. Lavaei
J. Lavaei
中科院分区:
--
文献类型:
--
作者:
Han Feng;Baturalp Yalcin;J. Lavaei

文献摘要

参考文献

被引文献

相似文献

我们研究了受大而稀疏扰动影响的线性时不变动力系统的识别,建模对抗性攻击或故障。在状态可测的假设下,我们通过求解约束套索型优化问题来建立系统矩阵恢复的充分必要条件。此外,只要扰动序列是小噪声值和大对抗值的组合,我们就提供估计误差的上限。我们的结果取决于套索文献中广泛使用的零空间属性,并且我们研究了该属性在什么条件下适用于线性时不变动力系统。最后,我们进一步研究特定概率模型的条件,并通过数值实验支持结果。
We study the identification of a linear time-invariant dynamical system affected by large-and-sparse disturbances modeling adversarial attacks or faults. Under the assumption that the states are measurable, we develop necessary and sufficient conditions for the recovery of the system matrices by solving a constrained lasso-type optimization problem. In addition, we provide an upper bound on the estimation error whenever the disturbance sequence is a combination of small noise values and large adversarial values. Our results depend on the null space property that has been widely used in the lasso literature, and we investigate under what conditions this property holds for linear time-invariant dynamical systems. Lastly, we further study the conditions for a specific probabilistic model and support the results with numerical experiments.
稳健二次回归的圆锥优化:确定性界限和统计分析
DOI: --
发表时间: 2018
期刊: Conference on Decision and Control
影响因子: --
作者:
Molybog, Igor;Madani, Ramtin;Lavaei, Javad
通讯作者: Lavaei, Javad
DOI: 10.1007/s10107-020-01590-2
发表时间: 2020-11
影响因子: 2.7
作者:
Igor Molybog;S. Sojoudi;J. Lavaei
通讯作者: Igor Molybog;S. Sojoudi;J. Lavaei
线性系统集合隶属度估计器的统计一致性
DOI: 10.1109/lcsys.2020.2990998
发表时间: 2020
影响因子: 3
作者:
Hespanhol, Pedro;Aswani, Anil
通讯作者: Aswani, Anil
对抗性攻击下动态系统的学习
DOI: --
发表时间: 2021
期刊: Proceedings of the IEEE Conference on Decision Control
影响因子: --
作者:
Feng, Han;Lavaei, Javad
通讯作者: Lavaei, Javad