A convergence analysis of the nonlinear Uzawa algorithm for saddle point problems

From MaRDI portal
Publication:2471291





A result of \textit{Z.-H. Cao} [Appl. Numer. Math 46, No. 2, 157--171 (2003; Zbl 1032.65029)] on saddle point problems with penalty terms is improved. To this end a condition on the inexact Uzawa algorithm is slightly relaxed.











This page was built for publication: A convergence analysis of the nonlinear Uzawa algorithm for saddle point problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2471291)