On n-Person Game Solutions and Convex Programs with Essentially Unconstrained Duals

A. Ben-Tal, A. Charnes, B. Golany

Research output: Contribution to journalArticlepeer-review

Abstract

A special class of the Charnes-Kortanek “convex nucleus” solutions is studied in which the minimized functional is separable (monotropic). This class includes Euclidean distance, Hellinger measure, Bose-Einstein entropic measure and the Charnes-Cooper relative entropic solutions. It is shown that the dual convex programs are “essentially” unconstrained and that a simple explicit inversion formula connects optimal primal and dual solutions.

Original languageEnglish
Pages (from-to)71-84
Number of pages14
JournalOptimization
Volume19
Issue number1
DOIs
StatePublished - 1 Jan 1988

Keywords

  • Characteristic Function Games
  • Convex Nucleus Solutions
  • Essentially Unconstrained Duals
  • Monotropic Convex Programs

ASJC Scopus subject areas

  • Control and Optimization
  • Management Science and Operations Research
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'On n-Person Game Solutions and Convex Programs with Essentially Unconstrained Duals'. Together they form a unique fingerprint.

Cite this