Answers are grounded in this paper and cite their sources.
This paper proposes an adaptive innovation for selecting regularization parameters with optimal features. Firstly, the least squares solution of the optimization problem is investigated by the Tikhonov regularization method to construct a quotient function with regularization parameters as independent variables. Secondly, for the quadratic programming theory, an adaptive innovation for selecting the optimal regularization parameters put forward by regularizing the relationship between the regularization and the function values. Finally, the proposed method is compared with traditional methods which are widely used through numerical simulation. The results show that the quotient function method can effectively overcome the ill-posed problem caused by the ill-posedness of the system matrix since it has certain anti-noise performance, and can theoretically obtains an approximate stable solution with better accuracy.
No takes yet. Share an insight, caveat, or question.
Feng et al. (2020) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: