Journal
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
Volume 33, Issue 3, Pages 1216-1227Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2020.3041364
Keywords
Computational modeling; Mathematical model; Neural networks; Control theory; Analytical models; Real-time systems; Numerical models; Control-theoretical techniques; perturbed nonstationary quadratic program (QP); recurrent neural dynamics; robustness theoretical analysis
Categories
Funding
- National Natural Science Foundation of China [61703189, 11561029, 61772493]
- Pioneer Hundred Talents Program of the Chinese Academy of Sciences
- National Key Research and Development Program of China [2017YFE0118900]
- Natural Science Foundation of Chongqing (China) [cstc2019jcyjjqX0013, cstc2020jcyj-zdxmX0028]
- Research and Development Foundation of Nanchong (China) [20YFZJ0018]
- CAS Light of West China Program
- Chongqing Key Laboratory of Mobile Communications Technology [cqupt-mct-202004]
- Ministry of Science and Higher Education of the Russian Federation as part of World-class Research Center program: Advanced Digital Technologies [075-15-2020903]
- Fundamental Research Funds for the Central Universities [lzujbky-2019-89, lzuxxxy-2019-tm11]
Ask authors/readers for more resources
This paper constructs a new recurrent neural dynamics model using control-theoretical techniques to tackle nonstationary quadratic programming problems, effectively breaking through the limitations of traditional models and demonstrating excellent convergence and robustness.
Recent decades have witnessed a trend that control-theoretical techniques are widely leveraged in various areas, e.g., design and analysis of computational models. Computational methods can be modeled as a controller and searching the equilibrium point of a dynamical system is identical to solving an algebraic equation. Thus, absorbing mature technologies in control theory and integrating it with neural dynamics models can lead to new achievements. This work makes progress along this direction by applying control-theoretical techniques to construct new recurrent neural dynamics for manipulating a perturbed nonstationary quadratic program (QP) with time-varying parameters considered. Specifically, to break the limitations of existing continuous-time models in handling nonstationary problems, a discrete recurrent neural dynamics model is proposed to robustly deal with noise. This work shows how iterative computational methods for solving nonstationary QP can be revisited, designed, and analyzed in a control framework. A modified Newton iteration model and an improved gradient-based neural dynamics are established by referring to the superior structural technology of the presented recurrent neural dynamics, where the chief breakthrough is their excellent convergence and robustness over the traditional models. Numerical experiments are conducted to show the eminence of the proposed models in solving perturbed nonstationary QP.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available