An Algorithm for Strictly Convex Quadratic Programming with Box Constraints

We propose an efficient algorithm for solving strictly convex quadratic programs with box constraints (i.e., lower and upper bounds on each variable). our algorithm is based on the active set and the Newton method. We repeatedly compute relevant inverse matrices efficiently and, starting from an initial feasible poillt, we find an optimal solution of the problem in finitely many steps. Our experimental results show that the new algorithm is practically efficient.