论文信息 - Regularized Decomposition and Augmented Lagrangian Decomposition for Angular Linear Programming Problems

Regularized Decomposition and Augmented Lagrangian Decomposition for Angular Linear Programming Problems

We present two new decomposition methods for large linear programming problems of angular structure. The first one is a special version of the regularized decomposition method and the second one is a decomposable version of the augmented Lagrangian method. For both methods we prove finite termination theorems and establish their duality.

A. Ruszczynski

[1] George B. Dantzig,et al. Decomposition Principle for Linear Programs , 1960 .

[2] R. Tyrrell Rockafellar,et al. Augmented Lagrangians and Applications of the Proximal Point Algorithm in Convex Programming , 1976, Math. Oper. Res..

[3] P. Shepherd,et al. Large scale systems theory and applications: Udine, Italy, 16–20 June 1976 , 1976 .

[4] R. Tapia. Diagonalized multiplier methods and quasi-Newton methods for constrained optimization , 1977 .

[5] N. Watanabe,et al. Decomposition in large system optimization using the method of multipliers , 1978 .

[6] Dimitri P. Bertsekas,et al. Constrained Optimization and Lagrange Multiplier Methods , 1982 .

[7] K. Kiwiel. Methods of Descent for Nondifferentiable Optimization , 1985 .

[8] Andrzej Ruszczynski,et al. A regularized decomposition method for minimizing a sum of polyhedral functions , 1986, Math. Program..

[9] R. Tyrrell Rockafellar,et al. Scenarios and Policy Aggregation in Optimization Under Uncertainty , 1991, Math. Oper. Res..