Reachability Analysis and Safety Verification of Neural Feedback Systems via Hybrid Zonotopes
Reachability Analysis and Safety Verification of Neural Feedback Systems via Hybrid Zonotopes
复制标题
通过混合区域的神经反馈系统的可达性分析和安全验证
DOI：
10.23919/acc55779.2023.10156417
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Xiangru Xu
中科院分区：
文献类型：
--
作者：
Yuhao Zhang;Xiangru Xu
通过引入其他二进制变量并具有一些独特的属性，可以使它们方便地表示非convex集，从而概括了受约束的分区。本文介绍了基于新型混合生物学的方法，用于神经反馈系统的可及性分析和安全验证。提出了算法来计算馈送前向神经网络的每一层的输入输出关系，以及神经反馈系统的确切可触及集。结果表明，恢复活化的馈送神经网络可以由混合发射型精确表示。此外，将足够且必要的条件作为混合企业线性程序制定，以证明神经反馈系统的轨迹是否可以避免不安全的区域。所提出的方法显示出一种制定的配方，为神经反馈系统的可触及集提供了最紧密的凸松弛。开发了可达集的复杂性降低技术，以平衡计算效率和近似精度。两个数值示例表明，与其他现有方法相比，所提出的方法的出色表现。
Hybrid zonotopes generalize constrained zonotopes by introducing additional binary variables and possess some unique properties that make them convenient to represent nonconvex sets. This paper presents novel hybrid zonotope-based methods for the reachability analysis and safety verification of neural feedback systems. Algorithms are proposed to compute the input-output relationship of each layer of a feed-forward neural network, as well as the exact reachable sets of neural feedback systems. It is shown that a ReLU-activated feed-forward neural network can be exactly represented by a hybrid zonotope. In addition, a sufficient and necessary condition is formulated as a mixed-integer linear program to certify whether the trajectories of a neural feedback system can avoid unsafe regions. The proposed approach is shown to yield a formulation that provides the tightest convex relaxation for the reachable sets of the neural feedback system. Complexity reduction techniques for the reachable sets are developed to balance the computation efficiency and approximation accuracy. Two numerical examples demonstrate the superior performance of the proposed approach compared to other existing methods.
影响因子：
6.8
作者：
Fazlyab, Mahyar;Morari, Manfred;Pappas, George J.
通讯作者：
Pappas, George J.
DOI：
10.1109/cdc42340.2020.9304296
发表时间：
2020-04
期刊：
2020 59th IEEE Conference on Decision and Control (CDC)
影响因子：
--
作者：
Haimin Hu;Mahyar Fazlyab;M. Morari;George Pappas
通讯作者：
Haimin Hu;Mahyar Fazlyab;M. Morari;George Pappas
影响因子：
2
作者：
Hoang-Dung Tran;Cai, Feiyang;Koutsoukos, Xenofon
通讯作者：
Koutsoukos, Xenofon