Browse · MathNet
PrintIndia_2017
India 2017 counting and probability
Problem
Define an -magic square to mean an square matrix of non-negative integers such that the sum of all the entries in each row and each column is for some . Also define an -permutation matrix to mean an square matrix of zeroes and ones such that every row and every column of the array contains exactly one 1. Show that every -magic square can be written as a sum of finitely many -permutation matrices.
Solution
Fix an . We will prove the following statement by Induction: : Every -magic square with common sum can be written as a sum of permutation matrices. For the base case, we verify that is true. Since all the entries are non-negative integers, the only way possible -magic squares with common sum 1 are exactly -permutation matrices. Hence, every such matrix can be written as a sum of -permutation matrix, i.e., itself. For the Induction Hypothesis, assume that the statement is true for some . Hence, every -magic square with common sum can be written as a sum of -permutation matrices. Consider any -magic square with common sum . We will prove that can be written as a sum of an -permutation matrix and an -magic square with common sum . Consider the graph defined as follows. Let be the rows of and let be the columns of . Consider . Each of the sets and are independent sets. For every , are connected by an edge if and only if , where is the entry of in the -th row and -th column. Observe that from our definition, is a bipartite graph. Consider some . By , we will denote the set of vertices which are adjacent to some vertex in . We will show that . For every edge having an endpoint in , label that edge with . Consider the 'sum' of all such edges. Clearly, for each , the edges incident with contribute to this total. Hence, the total sum is given by . Now these are edges are a subset of all edges with an endpoint in . Hence, if we consider a similar such sum for the set , we get the sum to be . As observed before, we see that . Hence, we get that . Since was any arbitrary subset of , we that this is true for all subsets of . Since the hypothesis for Hall's Matching Theorem is satisfied, an application of it implies that there exists a matching in the above bipartite graph . Since , we see that the matching is nothing but a bijective function where is defined to be the index of the element in which is matched with . Note that can also be thought of as a permutation of . We also know that . Since is a matching, we know that for all . Consider the permutation matrix where each entry is given by Also consider the matrix . Note that since , every entry of is non-negative. Also observe that since is a permutation matrix, the sum of entries in each row and each column reduces by 1. In effect, is still an -magic square with common sum . This completes the inductive step. Hence by Induction, we see that the statement is true for all . The proof is now complete.
Techniques
Matchings, Marriage Lemma, Tutte's theoremInduction / smoothing