Abstract
For integer , let be the number of rows of the largest all-0 or all-1 square submatrix of , minimized over all -matrices . Thus . But let us fix a matrix , and define to be the same, minimized over over all -matrices such that neither nor its complement (that is, change all 's to 's and vice versa) contains as a submatrix. It is known that , where are constants depending on . When can we take ? If so, then one of and its complement must be an acyclic matrix (that is, the corresponding bipartite graph is a forest). Korandi, Pach, and Tomon conjectured the converse, that is linear in for every acyclic matrix ; and they proved it for certain matrices with only two rows. Their conjecture remains open, but we show for every acyclic matrix ; and indeed there is a -submatrix that is either or .
Connections
Explore this paper’s authors, topics and related work.
Building this map preview
BZPEER is loading the nearby papers, people, topics and institutions for this page.
Reviews 0
Write a review
No reviews yet.
Discussion 0
Add a comment
No comments yet.