Robust standard gradient descent algorithm for ARX models using Aitken acceleration technique

Chen, Jing; Gan, Min; Zhu, Quanmin; Narayan, Pritesh; Liu, Yanjun

Abstract

A robust standard gradient descent (SGD) algorithm for ARX models using the Aitken acceleration method is developed. Considering that the SGD algorithm has slow convergence rates and is sensitive to the step size, a robust and accelerative SGD (RA-SGD) algorithm is derived. This algorithm is based on the Aitken acceleration method, and its convergence rate is improved from linear convergence to at least quadratic convergence in general. Furthermore, the RA-SGD algorithm is always convergent with no limitation of the step size. Both the convergence analysis and the simulation examples demonstrate that the presented algorithm is effective.

Citation

Chen, J., Gan, M., Zhu, Q., Narayan, P., & Liu, Y. (2022). Robust standard gradient descent algorithm for ARX models using Aitken acceleration technique. IEEE Transactions on Cybernetics, 52(9), 9646-9655. https://doi.org/10.1109/tcyb.2021.3063113

• Journal Article Type: Article
• Acceptance Date: Feb 24, 2021
• Online Publication Date: Mar 23, 2021
• Publication Date: 2022-09
• Deposit Date: Mar 25, 2021
• Publicly Available Date: Aug 11, 2021
• Journal: IEEE Transactions on Cybernetics
• Print ISSN: 2168-2267
• Electronic ISSN: 2168-2275
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 52
• Issue: 9
• Pages: 9646-9655
• DOI: https://doi.org/10.1109/tcyb.2021.3063113
• Keywords: Control and Systems Engineering; Human-Computer Interaction; Electrical and Electronic Engineering; Software; Information Systems; Computer Science Applications
• Public URL: https://uwe-repository.worktribe.com/output/7231652