Greedy Search Method for Separable Nonlinear Models Using Stage Aitken Gradient Descent and Least Squares Algorithms

• 成果类型: Article
• 署名作者: Chen, Jing; Mao, Yawen; Gan, Min; Wang, Dongqing; Zhu, Quanmin
• 署名单位: Jiangnan University; Qingdao University; Qingdao University
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2022.3214474
• 发表日期: 2023
• 页码: 5044-5051
• 关键词: Aitken acceleration technique; convergence rate; hierarchical identification algorithm; Parameter Estimation; separable nonlinear model

摘要

Aitken gradient descent (AGD) algorithm takes some advantages over the standard gradient descent and Newton methods: 1) can achieve at least quadratic convergence in general; 2) does not require the Hessian matrix inversion; 3) has less computational efforts. When using the AGD method for a considered model, the iterative function should be unchanging during all the iterations. This article proposes a hierarchical AGD algorithm for separable nonlinear models based on stage greedy method. The linear parameters are estimated using the least squares algorithm, and the nonlinear parameters are updated based on the AGD algorithm. Since the iterative function is changing at each iteration, a stage AGD algorithm is introduced. The convergence properties and simulation examples show effectiveness of the proposed algorithm.