Computationally efficient neural hybrid automaton framework for learning complex dynamics
Computationally efficient neural hybrid automaton framework for learning complex dynamics
复制标题
DOI：
10.1016/j.neucom.2023.126879
复制
发表时间：
2023-10
期刊：
影响因子：
6
通讯作者：
Tao Wang;Yejiang Yang;Weiming Xiang
中科院分区：
文献类型：
--
作者：
Tao Wang;Yejiang Yang;Weiming Xiang
本文提出了一种计算效率高且有效的动态系统数据驱动建模框架。提出的建模框架采用一组称为极限学习机（elm）的浅层神经网络来模拟局部系统行为，以及数据驱动的局部模型之间的推断转换，以建立神经混合自动机模型。首先，将采样的系统输入映射到相应的特征空间，获得数据驱动分区，然后通过一种新的数据驱动模式聚类过程定义神经混合自动机模型的过渡和不变量。然后，训练一组elm来近似局部动力学。学习过程集成了用于位置识别的分段数据合并过程和局部动力学建模过程。本文提出的神经混合自动机模型能够以较高的建模精度捕获复杂动力系统的行为，但在训练和验证等计算量大的任务中，计算复杂度显著降低，而这些任务传统上被认为是神经网络模型计算量大的任务。在此基础上，提出了一种基于区间分析的集值可达性分析方法，该方法计算效率高，是安全验证中常用的方法。最后，给出了极限环和人类手写运动建模的应用，以证明该方法的有效性和高效性。
This paper proposes a computationally efficient and effective data-driven modeling framework for dynamical systems. The proposed modeling framework employs a collection of shallow neural networks known as Extreme Learning Machines (ELMs) to model local system behaviors along with data-driven inferred transitions among local models to establish a neural hybrid automaton model. First, the sampled system inputs are mapped to the corresponding feature spaces to obtain data-driven partitions, which subsequently define the transitions and invariants of the neural hybrid automaton model through a novel data-driven mode clustering process. Then, a collection of ELMs are trained to approximate the local dynamics. The learning processes integrate a segmented data merging procedure for location identification and a local dynamics modeling process. The proposed neural hybrid automaton models can capture behaviors of complex dynamical systems with high modeling precision but significantly lower computational complexities in computationally expensive tasks such as training and verification, which are traditionally considered to be computationally expensive tasks for neural network models. A computationally efficient set-valued reachability analysis method which is commonly used in safety verification is then developed based on interval analysis and a novel Split and Combine process. Finally, applications to modeling the limit cycle and human handwritten motions are presented to show the effectiveness and efficiency of our approach.