Vector and matrix apportionment problems and separable convex integer optimization

  • Algorithms for the proportional rounding of a nonnegative vector, and for the biproportional rounding of a nonnegative matrix are discussed. Here we view vector and matrix rounding as special instances of a generic optimization problem that employs an additive version of the objective function of Gaffke and Pukelsheim (2007). The generic problem turns out to be a separable convex integer optimization problem, in which the linear equality constraints are given by a totally unimodular coefficient matrix. So, despite the integer restrictions of the variables, Fenchel duality applies. Our chief goal is to study the implied algorithmic consequences. We establish a general algorithm based on the primal optimization problem. Furthermore we show that the biproportional algorithm of Balinski and Demange (1989), when suitably generalized, derives from the dual optimization problem. Finally we comment on the shortcomings of the alternating scaling algorithm, a discrete variant of the well-knownAlgorithms for the proportional rounding of a nonnegative vector, and for the biproportional rounding of a nonnegative matrix are discussed. Here we view vector and matrix rounding as special instances of a generic optimization problem that employs an additive version of the objective function of Gaffke and Pukelsheim (2007). The generic problem turns out to be a separable convex integer optimization problem, in which the linear equality constraints are given by a totally unimodular coefficient matrix. So, despite the integer restrictions of the variables, Fenchel duality applies. Our chief goal is to study the implied algorithmic consequences. We establish a general algorithm based on the primal optimization problem. Furthermore we show that the biproportional algorithm of Balinski and Demange (1989), when suitably generalized, derives from the dual optimization problem. Finally we comment on the shortcomings of the alternating scaling algorithm, a discrete variant of the well-known Iterative Proportional Fitting procedure.show moreshow less

Download full text files

Export metadata

Statistics

Number of document requests

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Norbert GaffkeGND, Friedrich PukelsheimGND
URN:urn:nbn:de:bvb:384-opus4-4122
Frontdoor URLhttps://opus.bibliothek.uni-augsburg.de/opus4/511
Series (Serial Number):Preprints des Instituts für Mathematik der Universität Augsburg (2007-06)
Type:Preprint
Language:English
Publishing Institution:Universität Augsburg
Release Date:2007/05/30
Tag:totally unimodular matrix; elementary vector; Graver basis; convex programming duality; alternating maximization procedure
GND-Keyword:Optimierungsproblem; Ganzzahlige Optimierung; Dualitätstheorie; Unimodulare Matrix
Institutes:Mathematisch-Naturwissenschaftlich-Technische Fakultät
Mathematisch-Naturwissenschaftlich-Technische Fakultät / Institut für Mathematik
Mathematisch-Naturwissenschaftlich-Technische Fakultät / Institut für Mathematik / Lehrstuhl für Stochastik und ihre Anwendungen
Dewey Decimal Classification:5 Naturwissenschaften und Mathematik / 51 Mathematik / 510 Mathematik