On the convergence of the quasioptimality criterion for (iterated) Tikhonov regularization (Q998872)

From MaRDI portal
Revision as of 20:38, 30 January 2024 by Import240129110113 (talk | contribs) (Added link to MaRDI item.)
scientific article
Language Label Description Also known as
English
On the convergence of the quasioptimality criterion for (iterated) Tikhonov regularization
scientific article

    Statements

    On the convergence of the quasioptimality criterion for (iterated) Tikhonov regularization (English)
    0 references
    0 references
    0 references
    30 January 2009
    0 references
    The authors study the iterated Tikhonov regularization scheme \(x_{\alpha,n}^{\delta}=(T^*T+\alpha I)^{-1}(T^* y^{\delta}+\alpha x_{\alpha,n-1}^{\delta})\), \(x_{\alpha,0}^{\delta}=0\) for linear ill-posed problems \(Tx=y^{\delta}\), where \(\| y^{\delta}-y \| \leq \delta\), \(T\in L(X,Y)\), and \(X,Y\) are Hilbert spaces. For a special error free parameter choice rule \(\alpha=\alpha(y^{\delta})\), the authors derive conditions on \(y^{\delta}-y\) that ensure the convergence of the approximations \(x_{\alpha(y^{\delta}),n}^{\delta}\) to a solution of the unperturbed equation as \(\delta\to 0\). Estimates of the rate of convergence are also established.
    0 references
    quasi-optimality criterion
    0 references
    heuristic parameter selection
    0 references
    iterated Tikhonov regularisation
    0 references

    Identifiers