Separating doubly nonnegative and completely positive matrices

The cone of Completely Positive (CP) matrices can be used to exactly formulate a variety of NP-Hard optimization problems. A tractable relaxation for CP matrices is provided by the cone of Doubly Nonnegative (DNN) matrices; that is, matrices that are both positive semidefinite and componentwise nonnegative. A natural problem in the optimization setting is then to separate a given DNN but non-CP matrix from the cone of CP matrices. We describe two different constructions for such a separation that apply to 5 × 5 matrices that are DNN but non-CP. We also describe a generalization that applies to larger DNN but non-CP matrices having block structure. Computational results illustrate the applicability of these separation procedures to generate improved bounds on difficult problems.

[1]  Leslie J. Briggs,et al.  Principles of Instructional Design , 1974 .

[2]  Abraham Berman,et al.  5 × 5 Completely positive matrices , 2004 .

[3]  Fabio Tardella,et al.  New and old bounds for standard quadratic optimization: dominance, equivalence and incomparability , 2008, Math. Program..

[4]  Kurt M. Anstreicher,et al.  A note on “ 5 × 5 Completely positive matrices ” , 2010 .

[5]  Marvin C. McCallum,et al.  DEVELOPING PERFORMANCE-BASED ASSESSMENTS OF MARINER PROFICIENCY , 2001 .

[6]  Samuel Burer,et al.  Separation and relaxation for cones of quadratic forms , 2013, Math. Program..

[7]  Charles R. Johnson,et al.  The copositive completion problem , 2005 .

[8]  Jos F. Sturm,et al.  A Matlab toolbox for optimization over symmetric cones , 1999 .

[9]  Etienne de Klerk,et al.  Solving Standard Quadratic Optimization Problems via Linear, Semidefinite and Copositive Programming , 2002, J. Glob. Optim..

[10]  Etienne de Klerk,et al.  Approximation of the Stability Number of a Graph via Copositive Programming , 2002, SIAM J. Optim..

[11]  H. Väliaho,et al.  Almost copositive matrices , 1989 .

[12]  Yasutoshi Yajima,et al.  A Polyhedral Approach for Nonconvex Quadratic Programming Problems with Box Constraints , 1998, J. Glob. Optim..

[13]  Samuel Burer,et al.  Optimizing a polyhedral-semidefinite relaxation of completely positive programs , 2010, Math. Program. Comput..

[14]  Samuel Burer,et al.  On the copositive representation of binary and continuous nonconvex quadratic programs , 2009, Math. Program..

[15]  A. Berman,et al.  Completely Positive Matrices , 2003 .

[16]  Changqing Xu.  Completely positive matrices , 2004 .

[17]  Mirjam Dür,et al.  Interior points of the completely positive cone. , 2008 .

[18]  Francesco Barioli.  Completely positive matrices with a book-graph , 1998 .

[19]  Abraham Berman,et al.  Characterization of completely positive graphs , 1993, Discret. Math..

[20]  Marco Locatelli,et al.  Copositivity cuts for improving SDP bounds on the clique number , 2010, Math. Program..

[21]  A. N. Cockcroft,et al.  International convention on standards of training, certification and watchkeeping for seafarers, 1978, as amended , 2012 .

[22]  Adam N. Letchford,et al.  On Nonconvex Quadratic Programming with Box Constraints , 2009, SIAM J. Optim..

[23]  Samuel Burer,et al.  Computable representations for convex hulls of low-dimensional quadratic forms , 2010, Math. Program..

[24]  Mirjam Dür,et al.  An Adaptive Linear Approximation Algorithm for Copositive Programs , 2009, SIAM J. Optim..

[25]  Javier Peña,et al.  Computing the Stability Number of a Graph Via Linear and Semidefinite Programming , 2007, SIAM J. Optim..

[26]  Mirjam Dür,et al.  The difference between 5×5 doubly nonnegative and completely positive matrices , 2009 .

[27]  Merlin O'Neill,et al.  UNITED STATES COAST GUARD , 1950 .