Discrete Optimization Method Based on Grassmannian Parameterization in Multidimensional Dichotomic Data Structuring
P. V. Gracheva · 2011
A solution of the large computational time problem arising in multidimensional data struc� turing is addressed by employing algebraic properties of finite geometries. A vector parameterization of the Grassmannian Gr2 (k, n) is proposed which makes it possible to minimize the amount of mem� ory and reduce the number of operations required to solve the problem. An algorithm based on this parameterization and the Gray codes is constructed; the algorithm is suitable for parallel computation, which further reduces computation time. DOI: 10.3103/S1063454111040054