Divided powers are an algebraic gadget that emulate, in an arbitrary ring, the functions for all integers , together with the natural functional equations that they satisfy.
One of them is a nice binomial theorem without binomial coefficients : denoting by , one has
Another formula looks at what happens when one iterates the construction: if everything happens nicely, one has
In the development of the theory, the remark that comes then is the necessary observation that the coefficient is an integer, which is not obvious since it is written as a fraction. Two arguments are given to justify this fact.
- The formula which the authors claim can be proved by induction
- The observation that is the number of (unordered) partitions of a set with elements into subsets with elements.
The latter fact is a particular case of a more general counting question: if are integers and , then the number of (unordered) partitions of a set with elements into subsets of cardinalities is equal to where is the number of occurences of in the sequence .
The goal of this blog post is to present a combinatorial argument for the first equality, and an alternative expression of the second number as a product of integers. We also give a formal, group theoretical proof, that this quotient of factorials solves the given counting problem.
Counting partitions of given type
Let be a set with elements.
A partition of is a subset of
consisting of nonempty, pairwise disjoint, subsets whose union is .
Its *type* is the “multiset” of integers given by their cardinalities.
Because no is empty, the type is a multiset of nonzero integers.
It is slightly easier to authorize zero in this type;
then a partition has to be considered as a multiset of subsets of ,
still assumed pairwise disjoint, so that only the empty subset can be repeated.
The group of permutations of acts
on the set of partitions of .
This action preserves the type of a partition, so that
we get an action on the set on the set of partitions of type ,
of any multiset of integers . It is nonempty if and only if .
This action is transitive: Given two partitions
and with the same type ,
numbered so that for all ,
we just define a permutation of that maps the elements of to .
By the orbit-stabilizer theorem, the number of partitions of type
is equal to , where is the stabilizer
of any partition of type .
Since , one has .
On the other hand, the stabilizer of a partition
can be described as follows. By an element of this stabilizer,
the elements of are mapped to some set
which has the same cardinality as . In other words,
there is a morphism of groups from to the subgroup of the
symmetric group consisting of permutations
that preserve the cardinality. This subgroup is a product
of symmetric groups , for all ,
where is the number of occurrences of in .
Here, we omit the factor corresponding to because our morphism
doesnt' see it.
This morphism is surjective, because it has a section.
Given a permutation of that preserves
the cardinalities, we have essentially shown above how to construct
a permutation of that induces , and this permutation
belongs to . More precisely, if we number the elements
of as , we lift
to the permutation that maps to for all .
This makes sense since for all .
The kernel of this morphism
consists of permutations that fix each .
It is clearly equal to the product of the symmetric groups .
One thus has
,
as was to be shown.
A combinatorial interpretation of the first relation
As we said above, that relation
can be proved by induction, just playing with factorials,
but we want a combinatorial interpretation for it.
(Here we assume that .)
By the preceding paragraph,
the left hand side is the number of partitions of a set with elements
of type .
We may assume that this set is numbered .
Such a partition can be defined as follows.
First, we choose the subset that contains : this amounts to choosing
elements among , and there are
possibilities.
Now, we have to choose the subset that contains the smallest element
which is not in the first chosen subset.
There are elements to choose among the remaining ones,
hence possibilities.
And we go on in the same way, until we have chosen the required subsets.
(For the last one, there is course only one such choice, and
notice that the last factor is .)
An integral formula for the number of partitions of given type
Finally, we wish to give an alternating formula for the number
of partitions of given type of a set
that makes it clear that it is an integer.
Again, we can assume that the set is equal to
and that is a multiset of integers such that .
Let us write in an other way, ,
meaning that is repeated times, is repeated times, etc.
One has .
The idea is that we can first partition into a
subsets of cardinalities , ,… and then
subpartition these subsets: the first one into subsets with one element,
the second into subsets with two elements, etc.
The number of ordered partitions with , … elements
is the multinomial number
And the other choice is the product of integers as given in the preceding
section.
This gives
We can also write the fractions that appear in the product
as integers: