A new embedding for the augmented Lagrange method (Q2783493)

Different algorithms such as penalty, barrier, augmented Lagrangian and parametrical approaches are used in the solution of nonlinear optimization problems. The author develops a particular embedding for the augmented Lagrangian method following the line of \textit{D. Dentcheva, J. Guddat, J. J. Rückman} and \textit{K. Wendler} [Z. Oper. Res. 41, No. 2, 127-152 (1995; Zbl 0854.90127)]. She proves that for almost every fixed parameter the original optimization problem is regular in the sense of \textit{H. Th. Jongen, P. Jonker} and \textit{F. Twilt} [Math. Program. 34, 333-353 (1986; Zbl 0599.90114)]. Two examples are presented.