Blog Archive

Thursday, November 13, 2025

Grothendieck Rings and the Categorification of ℤ[M]

Let M be a monoid, view it as a discrete category D(M). Let Vectkf be the category of finite dimensional vector space over k.

Consider the following functor category C:=Funcf(D(M),Vectkf).

Here we only consider those functors with finite supports.

|{m∈M:F(m)≠0}|<∞

The functor category has a natural monoidal structure, which is given by

F⊗G(m):=⨁ab=mF(a)⊗kG(b)

Easy to see that

F≅G⟺∀m∈M,dim⁡F(m)=dim⁡G(m)

For a functor F∈C, we should define the dimension of F as:

dim⁡F:=∑m∈Mdim⁡F(m)⋅m∈Z[M]

Then the dim gives us an isomorphic between the Grothendieck Ring of C and Z[M].

dim:K0(C)→Z[M]

Hence C is a categorification of Z[M].

Now let us look at some combinatorics.

Let M=(N,+). Then Z[M]≅Z[x]. We know that ⋀(V) is a N graded vector space, and ⋀ maps direct sums to tensor products.

We have

⋀(kn)≅⨂i=1n⋀(k)

Take dim both sides we get

dim⁡⋀(kn)=∑i=0n(ni)xi=(1+x)n=dim⁡⨂i=1n⋀(k)

If we apply Totaldim:K0(C)→Z,[F]⟼∑m∈Mdim⁡F(m), we get

∑i=0n(ni)=2n

Also, we have

⋀(kn⊕km)≅⋀(kn)⊗k⋀(km)

Hence

⋀s(kn⊕km)≅⨁i+j=s⋀i(kn)⊗k⋀j(km)

Take the ordinary dimension both sides we get

(n+ms)=∑i+j=s(ni)(mj)

Equivalently we could consider

dimm:K0(C)→Z,F⟼dim⁡F(m)

 

No comments:

Post a Comment

Popular Posts