Dubins–Spanier theorems

The Dubins–Spanier theorems are several theorems in the theory of fair cake-cutting. They were published by Lester Dubins and Edwin Spanier in 1961.[1] Although the original motivation for these theorems is fair division, they are in fact general theorems in measure theory.

Setting edit

There is a set  , and a set   which is a sigma-algebra of subsets of  .

There are   partners. Every partner   has a personal value measure  . This function determines how much each subset of   is worth to that partner.

Let   a partition of   to   measurable sets:  . Define the matrix   as the following   matrix:

 

This matrix contains the valuations of all players to all pieces of the partition.

Let   be the collection of all such matrices (for the same value measures, the same  , and different partitions):

 

The Dubins–Spanier theorems deal with the topological properties of  .

Statements edit

If all value measures   are countably-additive and nonatomic, then:

This was already proved by Dvoretzky, Wald, and Wolfowitz.[2]

Corollaries edit

Consensus partition edit

A cake partition   to k pieces is called a consensus partition with weights   (also called exact division) if:

 

I.e, there is a consensus among all partners that the value of piece j is exactly  .

Suppose, from now on, that   are weights whose sum is 1:

 

and the value measures are normalized such that each partner values the entire cake as exactly 1:

 

The convexity part of the DS theorem implies that:[1]: 5 

If all value measures are countably-additive and nonatomic,
then a consensus partition exists.

PROOF: For every  , define a partition   as follows:

 
 

In the partition  , all partners value the  -th piece as 1 and all other pieces as 0. Hence, in the matrix  , there are ones on the  -th column and zeros everywhere else.

By convexity, there is a partition   such that:

 

In that matrix, the  -th column contains only the value  . This means that, in the partition  , all partners value the  -th piece as exactly  .

Note: this corollary confirms a previous assertion by Hugo Steinhaus. It also gives an affirmative answer to the problem of the Nile provided that there are only a finite number of flood heights.

Super-proportional division edit

A cake partition   to n pieces (one piece per partner) is called a super-proportional division with weights   if:

 

I.e, the piece allotted to partner   is strictly more valuable for him than what he deserves. The following statement is Dubins-Spanier Theorem on the existence of super-proportional division : 6 

Theorem — With these notations, let   be weights whose sum is 1, assume that the point   is an interior point of the (n-1)-dimensional simplex with coordinates pairwise different, and assume that the value measures   are normalized such that each partner values the entire cake as exactly 1 (i.e. they are non-atomic probability measures). If at least two of the measures   are not equal (  ), then super-proportional divisions exist.

The hypothesis that the value measures   are not identical is necessary. Otherwise, the sum   leads to a contradiction.

Namely, if all value measures are countably-additive and non-atomic, and if there are two partners   such that  , then a super-proportional division exists.I.e, the necessary condition is also sufficient.

Sketch of Proof edit

Suppose w.l.o.g. that  . Then there is some piece of the cake,  , such that  . Let   be the complement of  ; then  . This means that  . However,  . Hence, either   or  . Suppose w.l.o.g. that   and   are true.

Define the following partitions:

  •  : the partition that gives   to partner 1,   to partner 2, and nothing to all others.
  •   (for  ): the partition that gives the entire cake to partner   and nothing to all others.

Here, we are interested only in the diagonals of the matrices  , which represent the valuations of the partners to their own pieces:

  • In  , entry 1 is  , entry 2 is  , and the other entries are 0.
  • In   (for  ), entry   is 1 and the other entires are 0.

By convexity, for every set of weights   there is a partition   such that:

 

It is possible to select the weights   such that, in the diagonal of  , the entries are in the same ratios as the weights  . Since we assumed that  , it is possible to prove that  , so   is a super-proportional division.

Utilitarian-optimal division edit

A cake partition   to n pieces (one piece per partner) is called utilitarian-optimal if it maximizes the sum of values. I.e, it maximizes:

 

Utilitarian-optimal divisions do not always exist. For example, suppose   is the set of positive integers. There are two partners. Both value the entire set   as 1. Partner 1 assigns a positive value to every integer and partner 2 assigns zero value to every finite subset. From a utilitarian point of view, it is best to give partner 1 a large finite subset and give the remainder to partner 2. When the set given to partner 1 becomes larger and larger, the sum-of-values becomes closer and closer to 2, but it never approaches 2. So there is no utilitarian-optimal division.

The problem with the above example is that the value measure of partner 2 is finitely-additive but not countably-additive.

The compactness part of the DS theorem immediately implies that:[1]: 7 

If all value measures are countably-additive and nonatomic,
then a utilitarian-optimal division exists.

In this special case, non-atomicity is not required: if all value measures are countably-additive, then a utilitarian-optimal partition exists.[1]: 7 

Leximin-optimal division edit

A cake partition   to n pieces (one piece per partner) is called leximin-optimal with weights   if it maximizes the lexicographically-ordered vector of relative values. I.e, it maximizes the following vector:

 

where the partners are indexed such that:

 

A leximin-optimal partition maximizes the value of the poorest partner (relative to his weight); subject to that, it maximizes the value of the next-poorest partner (relative to his weight); etc.

The compactness part of the DS theorem immediately implies that:[1]: 8 

If all value measures are countably-additive and nonatomic,
then a leximin-optimal division exists.

Further developments edit

  • The leximin-optimality criterion, introduced by Dubins and Spanier, has been studied extensively later. In particular, in the problem of cake-cutting, it was studied by Marco Dall'Aglio.[3]

See also edit

References edit

  1. ^ a b c d e Dubins, Lester Eli; Spanier, Edwin Henry (1961). "How to Cut a Cake Fairly". The American Mathematical Monthly. 68 (1): 1–17. doi:10.2307/2311357. JSTOR 2311357.
  2. ^ Dvoretzky, A.; Wald, A.; Wolfowitz, J. (1951). "Relations among certain ranges of vector measures". Pacific Journal of Mathematics. 1: 59–74. doi:10.2140/pjm.1951.1.59.
  3. ^ Dall'Aglio, Marco (2001). "The Dubins–Spanier optimization problem in fair division theory". Journal of Computational and Applied Mathematics. 130 (1–2): 17–40. Bibcode:2001JCoAM.130...17D. doi:10.1016/S0377-0427(99)00393-3.
  4. ^ Neyman, J (1946). "Un théorèm dʼexistence". C. R. Acad. Sci. 222: 843–845.