A Dual-Feasible Active-Set Method for Quadratic Programming
We present a dual-feasible active-set method for convex quadratic programming. At each iteration of the algorithm, the dual variables are kept feasible with respect to the optimality conditions of the problem while allowing infeasibility in the primal variables. In addition, the method uses the Schur-complement method to solve KKT systems, allowing flexibility in the implementation of the linear algebra aspects of the method.
Tuesday, June 2, 2009
11:00AM AP&M 2402
Center for Computational Mathematics9500 Gilman Dr. #0112La Jolla, CA 92093-0112Tel: (858)534-9056