We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Science and Game Theory

Title: Zero-Sum Games and Linear Programming Duality

Abstract: The minimax theorem for zero-sum games is easily proved from the strong duality theorem of linear programming. For the converse direction, the standard proof by Dantzig (1951) is massively incomplete, as we argue in this article. We explain and combine classical theorems about solving linear equations with nonnegative variables to give a correct alternative proof.
Subjects: Computer Science and Game Theory (cs.GT); Optimization and Control (math.OC)
MSC classes: 91A05, 90C05
ACM classes: G.2
Cite as: arXiv:2205.11196 [cs.GT]
  (or arXiv:2205.11196v1 [cs.GT] for this version)

Submission history

From: Bernhard von Stengel [view email]
[v1] Mon, 23 May 2022 11:06:04 GMT (25kb)

Link back to: arXiv, form interface, contact.