| A PREDICTOR-CORRECTOR INTERIOR-POINT ALGORITHM FOR CONVEX QUADRATIC PROGRAMMING |
| |
| 作者姓名： | 梁昔明 钱积新 |
| |
| 作者单位： | 梁昔明(College of Information Science ＆ Engineering, Central South University, Changsha 410083);钱积新(Institute of Systems Engineering, National Lab of Industrial Control Technology, Zhejiang University, Hangzhou 310027)
|
| |
| 摘 要： | The simplified Newton method, at the expense of fast convergence, reduces the work required by Newton method by reusing the initial Jacobian matrix. The composite Newton method attempts to balance the trade-off between expense and fast convergence by composing one Newton step with one simplified Newton step. Recently, Mehrotra suggested a predictor-corrector variant of primal-dual interior point method for linear programming. It is currently the interior-point method of the choice for linear programming. In this work we propose a predictor-corrector interior-point algorithm for convex quadratic programming. It is proved that the algorithm is equivalent to a level-1 perturbed composite Newton method. Computations in the algorithm do not require that the initial primal and dual points be feasible. Numerical experiments are made.
|
A PREDICTOR-CORRECTOR INTERIOR-POINT ALGORITHM FOR CONVEX QUADRATIC PROGRAMMING |
| |
| Liang XimingQian Jixin.A PREDICTOR-CORRECTOR INTERIOR-POINT ALGORITHM FOR CONVEX QUADRATIC PROGRAMMING[J].Numerical Mathematics A Journal of Chinese Universities English Series,2002,11(1):52-62. |
| |
| Authors: | Liang XimingQian Jixin |
| |
| Institution: | 1. College of Information Science ＆ Engineering, Central South University, Changsha 410083
2. Institute of Systems Engineering, National Lab of Industrial Control Technology, Zhejiang University, Hangzhou 310027 |
| |
| Abstract: | The simplified Newton method, at the expense of fast convergence, reduces the work required by Newton method by reusing the initial Jacobian matrix. The composite Newton method attempts to balance the trade-off between expense and fast convergence by composing one Newton step with one simplified Newton step. Recently, Mehrotra suggested a predictor-corrector variant of primal-dual interior point method for linear programming. It is currently the interior-point method of the choice for linear programming. In this work we propose a predictor-corrector interior-point algorithm for convex quadratic programming. It is proved that the algorithm is equivalent to a level-1 perturbed composite Newton method. Computations in the algorithm do not require that the initial primal and dual points be feasible. Numerical experiments are made. |
| |
| Keywords: | convex quadratic programming interior-point methods predictor-corrector algorithms numerical experiments |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
