Implementation of Balanced Matrix Recognition Algorithms

Author(s): Kraaij, H.J.L. (2023)

Abstract:
A {0,±1} matrix is balanced if it does not contain a square submatrix with two nonzero elements per row and column in which the sum of all entries is 2 modulo 4. G. Zambelli provided a polynomial algorithm to test balancedness of a matrix. This paper discusses the algorithm and its implementation, and compares it against a naive approach using an exponential algorithm. The running time of the polynomial algorithm is O(n^9) for {0, 1} matrices and O(n^11) for {0,±1} matrices, where n is the number of rows and columns of the input matrix.

Document(s):

Kraaij_BA_EEMCS.pdf