A Newton-bracketing method for a simple conic optimization problem

robust numerical algorithms conic relaxations Newton-bracketing method nonconvex quadratic optimization problems secant-bracketing method for generating valid bounds

Mathematics Subject Classification ID

Quadratic programming (90C20) Convex programming (90C25) Semidefinite programming (90C22)

Abstract: For the Lagrangian-DNN relaxation of quadratic optimization problems (QOPs), we propose a Newton-bracketing method to improve the performance of the bisection-projection method implemented in BBCPOP [to appear in ACM Tran. Softw., 2019]. The relaxation problem is converted into the problem of finding the largest zero

y^{*}

of a continuously differentiable (except at

y^{*}

) convex function

g : m a t h b b R i g h t a r r o w m a t h b b R

such that

g (y) = 0

if

y l e q y^{*}

and

g (y) > 0

otherwise. In theory, the method generates lower and upper bounds of

y^{*}

both converging to

y^{*}

. Their convergence is quadratic if the right derivative of

g

at

y^{*}

is positive. Accurate computation of

g^{'} (y)

is necessary for the robustness of the method, but it is difficult to achieve in practice. As an alternative, we present a secant-bracketing method. We demonstrate that the method improves the quality of the lower bounds obtained by BBCPOP and SDPNAL+ for binary QOP instances from BIQMAC. Moreover, new lower bounds for the unknown optimal values of large scale QAP instances from QAPLIB are reported.

Recommendations

Cites work

Cited in

(8)

Describes a project that uses

Uses Software

This page was built for publication: A Newton-bracketing method for a simple conic optimization problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4999334)