Adaptive constraint reduction for convex quadratic programming
- 229 Downloads
We propose an adaptive, constraint-reduced, primal-dual interior-point algorithm for convex quadratic programming with many more inequality constraints than variables. We reduce the computational effort by assembling, instead of the exact normal-equation matrix, an approximate matrix from a well chosen index set which includes indices of constraints that seem to be most critical. Starting with a large portion of the constraints, our proposed scheme excludes more unnecessary constraints at later iterations. We provide proofs for the global convergence and the quadratic local convergence rate of an affine-scaling variant. Numerical experiments on random problems, on a data-fitting problem, and on a problem in array pattern synthesis show the effectiveness of the constraint reduction in decreasing the time per iteration without significantly affecting the number of iterations. We note that a similar constraint-reduction approach can be applied to algorithms of Mehrotra’s predictor-corrector type, although no convergence theory is supplied.
KeywordsConvex quadratic programming Constraint reduction Column generation Primal-dual interior-point method
Unable to display preview. Download preview PDF.
- 2.Atkinson, K.E.: An Introduction to Numerical Analysis. Wiley, Washington (2005) Google Scholar
- 6.Dantzig, G.B., Ye, Y.: A build–up interior method for linear programming: Affine scaling form. Technical report, University of Iowa, Iowa City, IA 52242, USA (July 1991) Google Scholar
- 9.Hertog, D., Roos, C., Terlaky, T.: A build-up variant of the path-following method for LP. Technical Report DUT-TWI-91-47, Delft University of Technology, Delft, The Netherlands (1991) Google Scholar
- 11.Higham, N.J.: Analysis of the Cholesky decomposition of a semi-definite matrix. In: Cox, M.G., Hammarling, S.J. (eds.) Reliable Numerical Computation, Walton Street, Oxford OX2 6DP, UK, 1990, pp. 161–185. Oxford University Press, Oxford (1990) Google Scholar
- 13.Jung, J.H.: Adaptive constraint reduction for convex quadratic programming and training support vector machines. PhD thesis, University of Maryland (2008). Available at http://hdl.handle.net/1903/8020
- 18.Nocedal, J., Wright, S.J.: Numerical Optimization. Springer, Berlin (2000) Google Scholar
- 21.Schölkopf, B., Smola, A.J.: Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT Press, Cambridge (2001) Google Scholar
- 22.Tikhonov, A.N., Arsenin, V.Y.: Solutions of Ill-Posed Problems. Wiley, Washington (2005) Google Scholar
- 23.Tits, A.L., Zhou, J.L.: A simple, quadratically convergent interior point algorithm for linear programming and convex quadratic programming. In: Hager, W.W., Hearn, D.W., Pardalos, P.M. (eds.) Large Scale Optimization: State of the Art, pp. 411–427. Kluwer Academic, Dordrecht (1994) CrossRefGoogle Scholar
- 28.Winternitz, L., Nicholls, S.O., Tits, A., O’Leary, D.: A constraint reduced variant of Mehrotra’s Predictor-Corrector Algorithm, 2007. Submitted for publication. http://www.optimization-online.org/DB_FILE/2007/07/1734.pdf
- 33.Zhang, Y.: Solving large–scale linear programs by interior–point methods under the MATLAB environment. Technical Report 96–01, Dept. of Mathematics and Statistics, Univ. of Maryland Baltimore County (1996) Google Scholar