The complexity of computing the minimum rank of a sign pattern matrix

We show that computing the minimum rank of a sign pattern matrix is NP hard. Our proof is based on a simple but useful connection between minimum ranks of sign pattern matrices and the stretchability problem for pseudolines arrangements. In fact, our hardness result shows that it is already hard to determine if the minimum rank of a sign pattern matrix is $\le 3$. We complement this by giving a polynomial time algorithm for determining if a given sign pattern matrix has minimum rank $\le 2$. Our result answers one of the open problems from Linial et al. [Combinatorica, 27(4):439–463, 2007].