Optimization Online


New lower bounds and asymptotics for the cp-rank

Immanuel M. Bomze (immanuel.bomze***at***univie.ac.at)
Werner Schachinger (werner.schachinger***at***univie.ac.at)
Reinhard Ullriche (reinhard.ullrich***at***univie.ac.at)

Abstract: Let $p_n$ denote the largest possible cp-rank of an $n\times n$ completely positive matrix. This matrix parameter has its significance both in theory and applications, as it sheds light on the geometry and structure of the solution set of hard optimization problems in their completely positive formulation. Known bounds for $p_n$ are $s_n=\binom{n+1}2-4$, the current best upper bound, and the Drew-Johnson-Loewy (DJL) lower bound $d_n=\big\lfloor\frac{n^2}4\big\rfloor$. The famous DJL conjecture (1994) states that $p_n=d_n$. Here we show $\corr{p_n=\frac {n^2}2 +{\mathcal O}\big(n^{3/2}\big) = 2d_n+{\mathcal O}\big(n^{3/2}\big)}$, and construct counterexamples to the DJL conjecture for all $n\ge {12}$

Keywords: copositive optimization, completely positive matrices, nonnegative factorization

Category 1: Linear, Cone and Semidefinite Programming (Other )

Citation: Preprint NI14048-POP, Isaac Newton Institute for Mathematical Sciences, University of Cambridge UK, submitted

Download: [PDF]

Entry Submitted: 06/13/2014
Entry Accepted: 06/13/2014
Entry Last Modified: 06/18/2014

Modify/Update this entry

  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository


Coordinator's Board
Classification Scheme
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society