Behavior of DCA sequences for solving the trust-region subproblem
- 150 Downloads
From our results it follows that any DCA sequence for solving the trust-region subproblem (see Pham Dinh and Le Thi, in SIAM J Optim 8:476–505, 1998) is convergent provided that the basic matrix of the problem is nonsingular and it does not have multiple negative eigenvalues. Besides, under this additional assumption, there exists such an open set Ω containing the global minimizers and the unique local-nonglobal minimizer (if such exists) that any DCA sequence with the initial point from Ω is contained in the set and converges to a global minimizer or the local-nonglobal minimizer. Various examples are given to illustrate the limiting behavior and stability of the DCA sequences. Structure of the KKT point set of the trust-region subproblem is also analyzed.
KeywordsTrust-region subproblem DC algorithm DCA sequence Limiting behavior and stability Multiplicity of nonpositive eigenvalue
Mathematics Subject Classification (2000)65K05 65K10 90C30 90C35
Unable to display preview. Download preview PDF.
- 1.Conn, A.R., Gould, N.I.M., Toint, P.L.: Trust-region methods MPS-SIAM series on optimization, Philadelphia (2008)Google Scholar
- 7.Le Thi H.A., Pham Dinh T., Ngai H.V.: Characterizations of minimizers of a quadratic function with quadratic constraints (manuscript) (2006)Google Scholar
- 8.Lee G.M., Tam N.N., Yen N.D.: Quadratic Programming and Affine Variational Inequalities: A Qualitative Study, Series: “Nonconvex Optimization and its Applications”, vol. 78, Springer (2005)Google Scholar
- 9.Lee G.M., Tam N.N., Yen N.D.: Stability of linear-quadratic minimization over Euclidean balls, Institute of Mathematics, Vietnamese Academy of Science and Technology, E-Preprint No. 2007/10/02 (submitted)Google Scholar
- 14.Pardalos P.M., Resende M.: Handbook of Applied Optimization. Oxford University Press, Oxford (2002)Google Scholar
- 15.Pham Dinh T., Le Thi H.A.: Lagrangian stability and global optimality on nonconvex quadratic minimization over Euclidean balls and spheres. J. Convex Anal. 2, 263–276 (1995)Google Scholar
- 17.Pham Dinh T., Le Thi H.A.: Convex analysis approach to DC programming: theory, algorithms and applications. Acta Math. Vietnam 22, 289–355 (1997)Google Scholar