We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Optimization and Control

Title: An Optimality Gap Test for a Semidefinite Relaxation of a Quadratic Program with Two Quadratic Constraints

Abstract: We propose a necessary and sufficient test to determine whether a solution for a general quadratic program with two quadratic constraints (QC2QP) can be computed from that of a specific convex semidefinite relaxation, in which case we say that there is no optimality gap. Originally intended to solve a nonconvex optimal control problem, we consider the case in which the cost and both constraints of the QC2QP may be nonconvex. We obtained our test, which also ascertains when strong duality holds, by generalizing a closely-related method by Ai and Zhang. An extension was necessary because, while the method proposed by Ai and Zhang also allows for two quadratic constraints, it requires that at least one is strictly convex. In order to illustrate the usefulness of our test, we applied it to two examples that do not satisfy the assumptions required by prior methods. Our test guarantees that there is no optimality gap for the first example---a solution is also computed from the relaxation---and we used it to establish that an optimality gap exists in the second. We also verified using the test in a numerical experiment that there is no optimality gap in most instances of a set of randomly generated QC2QP, indicating that our method is likely to be useful in applications other than that of our original motivation.
Subjects: Optimization and Control (math.OC)
Cite as: arXiv:1907.02989 [math.OC]
  (or arXiv:1907.02989v2 [math.OC] for this version)

Submission history

From: Sheng Cheng [view email]
[v1] Fri, 5 Jul 2019 18:26:49 GMT (176kb)
[v2] Thu, 16 Jan 2020 13:51:54 GMT (176kb)

Link back to: arXiv, form interface, contact.