The verification of a local minimizer of a general (i.e., nonconvex) quadratic program is in general an NP-hard problem. The difficulty concerns the optimality of certain points (which we call dead points) at which the first-order necessary conditions for optimality are satisfied, but strict complementarity does not hold. One important class of methods for solving general quadratic prcgrammirg problems are called inertia-controlling quadratic programming (ICQP) methods. We derive a

more »

... scheme for proceeding at a dead point that is appropridte for a general ICQP method. The verification of a local minimizer of a general (ie, nonconvex) quadratic program is in general an NP-hard problem The difficulty concerns the optimality of certain points (which we call dead points) at which the first-order necessary conditions for optimality are satisfied, but strict complementarity does not hold One important call of methods for solving general quadratic programming problems are called zntrrtza-controllzng quadratzc programymnng (ICQP) methods. We derive a computational scheme for proceeding at a dead point that is appropriate for a general ICQP method SUCUNIIY CL-AIPlrCAV~OV 'u*** PAOl[t'll.