Solution of a tropical optimization problem using matrix sparcification

Результат исследований: Материалы конференцийтезисырецензирование

1 Загрузки (Pure)


A multidimensional optimization problem, which arises in various applications in the form of minimization of span seminorm is considered in the framework of tropical (idempotent) mathematics. The problem is formulated to minimize a nonlinear function, which is defined on vectors over an idempotent semifield, given by a matrix, and calculated using multiplicative conjugate transposition. To solve the problem, we apply and develop methods of tropical optimization proposed and investigated in [1-5]. First, we find the minimum value of the objective function, and give a partial solution as a subset of vectors represented in an explicit form. We characterize all solutions to the problem by a system of simultaneous vector equation and inequality, and use this characterization to investigate properties of the solution set, which, in particular, turns out to be closed under vector addition and scalar multiplication. Furthermore, a matrix sparsification technique is developed to drop, without affecting the solution of the problem, those entries in the matrix which are below prescribed threshold values. By combining this technique with the above characterization, the previous partial solution is extended to a wider solution subset, and then to a complete solution described as a family of subsets. Finally, we offer a backtracking procedure that generates all members of the family, and derive an explicit representation for the complete solution. Numerical examples and graphical illustrations of the results are also presented.

[1] Krivulin, N. A constrained tropical optimization problem: Complete solution and application example. In G. L. Litvinov and S. N. Sergeev (eds.) Tropical and Idempotent Mathematics and Applications, Contemp. Math., vol. 616, Providence, RI: AMS, 2014, pp. 163–177.
[2] Krivulin, N. Complete solution of a constrained tropical optimization problem with application to location analysis. In P. H¨ofner, P. Jipsen, W. Kahl, M. E. M¨uller (eds.) Relational and Algebraic Methods in Computer Science, Lecture Notes in Computer Science, vol. 8428, Cham: Springer, 2014, pp. 362–378.
[3] Krivulin, N. Tropical optimization problems. In L. A. Petrosyan, J. V. Romanovsky, and D. W. K. Yeung (eds.) Advances in Economics and Optimization: Collected Scientific Studies Dedicated to the Memory of L. V. Kantorovich, New York: Nova Sci. Publ., 2014, pp. 195–214.
[4] Krivulin, N. A multidimensional tropical optimization problem with nonlinear objective function and linear constraints. Optimization, vol. 64, no. 5, pp. 1107–1129, 2015.
Язык оригиналаанглийский
СостояниеОпубликовано - 2015
Событие4th International Conference on Matrix Methods in Mathematics and Applications - Skolkovo Institute of Science and Technology, Moscow, Российская Федерация
Продолжительность: 24 авг 201528 авг 2015


конференция4th International Conference on Matrix Methods in Mathematics and Applications
Сокращенный заголовокMMMA-2015
Страна/TерриторияРоссийская Федерация
Адрес в сети Интернет

Предметные области Scopus

  • Теория оптимизации
  • Алгебра и теория чисел


Подробные сведения о темах исследования «Solution of a tropical optimization problem using matrix sparcification». Вместе они формируют уникальный семантический отпечаток (fingerprint).