From: Fred Galvin Subject: Re: Stable marriages problem - looking for a proof Date: Thu, 5 Aug 1999 11:57:11 -0500 Newsgroups: sci.math Keywords: references On 5 Aug 1999, Remco Gerlich wrote: > For a lab session, I have to implement a program that finds a solution > for "the stable marriages problem". I'm thinking of using a simple > generate-and-test approach, but the text mentions that "this is always > possible, by a theorem from graph theory." > > I'd like to see the theorem and the proof, since it may suggest > a more efficient approach for my program. > > The stable marriages problem: given a group of n males and n females, > where each has a list of preferences for the members of the opposite > sex. Make n male-female couples that marry, where each marriage is > stable. A marriage is unstable if there is a couple in the group that > isn't married, and that has each other higher on their preferences > list than their spouses. > > Does anyone know which theorem from graph theory proves that > this is always possible, and if the proof is online somewhere? > Or a well-known book that has it? D. Gale and L. S. Shapley, College admissions and the stability of marriage, American Mathematical Monthly 69 (1962), 9-15. Dan Gusfield and Robert W. Irving, _The Stable Marriage Problem: Structures and Algorithms_, MIT Press, 1989. Donald E. Knuth, _Stable Marriage and Its Relation to Other Combinatorial Problems_, American Mathematical Society, 1997. Douglas B. West, _Introduction to Graph Theory_, Prentice-Hall, 1996. (Section 3.2, under "Stable Matchings")