English

Restricted completion of sparse partial Latin squares

Combinatorics 2019-08-15 v1 Discrete Mathematics

Abstract

An n×nn \times n partial Latin square PP is called α\alpha-dense if each row and column has at most αn\alpha n non-empty cells and each symbol occurs at most αn\alpha n times in PP. An n×nn \times n array AA where each cell contains a subset of {1,,n}\{1,\dots, n\} is a (βn,βn,βn)(\beta n, \beta n, \beta n)-array if each symbol occurs at most βn\beta n times in each row and column and each cell contains a set of size at most βn\beta n. Combining the notions of completing partial Latin squares and avoiding arrays, we prove that there are constants α,β>0\alpha, \beta > 0 such that, for every positive integer nn, if PP is an α\alpha-dense n×nn \times n partial Latin square, AA is an n×nn \times n (βn,βn,βn)(\beta n, \beta n, \beta n)-array, and no cell of PP contains a symbol that appears in the corresponding cell of AA, then there is a completion of PP that avoids AA; that is, there is a Latin square LL that agrees with PP on every non-empty cell of PP, and, for each i,ji,j satisfying 1i,jn1 \leq i,j \leq n, the symbol in position (i,j)(i,j) in LL does not appear in the corresponding cell of AA.

Cite

@article{arxiv.1608.07383,
  title  = {Restricted completion of sparse partial Latin squares},
  author = {Lina J. Andrén and Carl Johan Casselgren and Klas Markström},
  journal= {arXiv preprint arXiv:1608.07383},
  year   = {2019}
}

Comments

19 pages

R2 v1 2026-06-22T15:31:41.216Z