Standard smooth algebras #
In this file we define standard smooth algebras. For this we introduce
the notion of a PreSubmersivePresentation
. This is a presentation P
that has
fewer relations than generators. More precisely there exists an injective map from σ
to P.ι
. To such a presentation we may associate a jacobian. P
is then a submersive
presentation, if its jacobian is invertible.
Finally, a standard smooth algebra is an algebra that admits a submersive presentation.
While every standard smooth algebra is smooth, the converse does not hold. But if S
is R
-smooth,
then S
is R
-standard smooth locally on S
, i.e. there exists a set { t }
of S
that
generates the unit ideal, such that Sₜ
is R
-standard smooth for every t
(TODO, see below).
Main definitions #
All of these are in the Algebra
namespace. Let S
be an R
-algebra.
PreSubmersivePresentation
: APresentation
ofS
asR
-algebra, equipped with an injective mapP.map
fromσ
toP.vars
. This map is used to define the differential of a presubmersive presentation.
For a presubmersive presentation P
of S
over R
we make the following definitions:
PreSubmersivePresentation.differential
: A linear endomorphism ofσ → P.Ring
sending thej
-th standard basis vector, corresponding to thej
-th relation, to the vector of partial derivatives ofP.relation j
with respect to the coordinatesP.map i
fori : σ
.PreSubmersivePresentation.jacobian
: The determinant ofP.differential
.PreSubmersivePresentation.jacobiMatrix
: Ifσ
has aFintype
instance, we may form the matrix corresponding toP.differential
. Its determinant isP.jacobian
.SubmersivePresentation
: A submersive presentation is a finite, presubmersive presentationP
with inS
invertible jacobian.
Furthermore, for algebras we define:
Algebra.IsStandardSmooth
:S
isR
-standard smooth ifS
admits a submersiveR
-presentation.Algebra.IsStandardSmooth.relativeDimension
: IfS
isR
-standard smooth this is the dimension of an arbitrary submersiveR
-presentation ofS
. This is independent of the choice of the presentation (TODO, see below).Algebra.IsStandardSmoothOfRelativeDimension n
:S
isR
-standard smooth of relative dimensionn
if it admits a submersiveR
-presentation of dimensionn
.
Finally, for ring homomorphisms we define:
RingHom.IsStandardSmooth
: A ring homomorphismR →+* S
is standard smooth ifS
is standard smooth asR
-algebra.RingHom.IsStandardSmoothOfRelativeDimension n
: A ring homomorphismR →+* S
is standard smooth of relative dimensionn
ifS
is standard smooth of relative dimensionn
asR
-algebra.
TODO #
- Show that the module of Kaehler differentials of a standard smooth
R
-algebraS
of relative dimensionn
isS
-free of rankn
. In particular this shows that the relative dimension is independent of the choice of the standard smooth presentation. - Show that standard smooth algebras are smooth. This relies on the computation of the module of Kaehler differentials.
- Show that locally on the target, smooth algebras are standard smooth.
Implementation details #
Standard smooth algebras and ring homomorphisms feature 4 universe levels: The universe levels of the rings involved and the universe levels of the types of the variables and relations.
Notes #
This contribution was created as part of the AIM workshop "Formalizing algebraic geometry" in June 2024.
A PreSubmersivePresentation
of an R
-algebra S
is a Presentation
with relations equipped with an injective map : relations → vars
.
This map determines how the differential of P
is constructed. See
PreSubmersivePresentation.differential
for details.
- val : ι → S
- σ' : S → MvPolynomial ι R
- algebra : Algebra (MvPolynomial ι R) S
- map : σ → ι
A map from the relations type to the variables type. Used to compute the differential.
- map_inj : Function.Injective self.map
Instances For
The standard basis of σ → P.ring
.
Equations
- P.basis = Pi.basisFun P.Ring σ
Instances For
The differential of a P : PreSubmersivePresentation
is a P.Ring
-linear map on
σ → P.Ring
:
The j
-th standard basis vector, corresponding to the j
-th relation of P
, is mapped
to the vector of partial derivatives of P.relation j
with respect
to the coordinates P.map i
for all i : σ
.
The determinant of this map is the jacobian of P
used to define when a PreSubmersivePresentation
is submersive. See PreSubmersivePresentation.jacobian
.
Equations
- P.differential = (P.basis.constr P.Ring) fun (j i : σ) => (MvPolynomial.pderiv (P.map i)) (P.relation j)
Instances For
PreSubmersivePresentation.differential
pushed forward to S
via aeval P.val
.
Equations
- P.aevalDifferential = ((Pi.basisFun S σ).constr S) fun (j i : σ) => (MvPolynomial.aeval P.val) ((MvPolynomial.pderiv (P.map i)) (P.relation j))
Instances For
The jacobian of a P : PreSubmersivePresentation
is the determinant
of P.differential
viewed as element of S
.
Equations
- P.jacobian = (algebraMap P.Ring S) (LinearMap.det P.differential)
Instances For
If σ
has a Fintype
and DecidableEq
instance, the differential of P
can be expressed in matrix form.
Equations
- P.jacobiMatrix = (LinearMap.toMatrix P.basis P.basis) P.differential
Instances For
If algebraMap R S
is bijective, the empty generators are a pre-submersive
presentation with no relations.
Equations
- One or more equations did not get rendered due to their size.
Instances For
If S
is the localization of R
at r
, this is the canonical submersive presentation
of S
as R
-algebra.
Equations
- Algebra.PreSubmersivePresentation.localizationAway S r = { toPresentation := Algebra.Presentation.localizationAway S r, map := fun (x : Unit) => (), map_inj := ⋯ }
Instances For
Given an R
-algebra S
and an S
-algebra T
with pre-submersive presentations,
this is the canonical pre-submersive presentation of T
as an R
-algebra.
Equations
Instances For
The dimension of the composition of two finite submersive presentations is the sum of the dimensions.
Jacobian of composition #
Let S
be an R
-algebra and T
be an S
-algebra with presentations P
and Q
respectively.
In this section we compute the jacobian of the composition of Q
and P
to be
the product of the jacobians. For this we use a block decomposition of the jacobi matrix and show
that the upper-right block vanishes, the upper-left block has determinant jacobian of Q
and
the lower-right block has determinant jacobian of P
.
The jacobian of the composition of presentations is the product of the jacobians.
If P
is a pre-submersive presentation of S
over R
and T
is an R
-algebra, we
obtain a natural pre-submersive presentation of T ⊗[R] S
over T
.
Equations
- Algebra.PreSubmersivePresentation.baseChange T P = { toPresentation := Algebra.Presentation.baseChange T P.toPresentation, map := P.map, map_inj := ⋯ }
Instances For
Given a pre-submersive presentation P
and equivalences ι' ≃ ι
and
σ' ≃ σ
, this is the induced pre-sumbersive presentation with variables indexed
by ι
and relations indexed by `κ
Equations
Instances For
A PreSubmersivePresentation
is submersive if its jacobian is a unit in S
and the presentation is finite.
- val : ι → S
- σ' : S → MvPolynomial ι R
- algebra : Algebra (MvPolynomial ι R) S
- map : σ → ι
- map_inj : Function.Injective self.map
Instances For
If algebraMap R S
is bijective, the empty generators are a submersive
presentation with no relations.
Equations
- Algebra.SubmersivePresentation.ofBijectiveAlgebraMap h = { toPreSubmersivePresentation := Algebra.PreSubmersivePresentation.ofBijectiveAlgebraMap h, jacobian_isUnit := ⋯ }
Instances For
The canonical submersive R
-presentation of R
with no generators and no relations.
Equations
Instances For
Given an R
-algebra S
and an S
-algebra T
with submersive presentations,
this is the canonical submersive presentation of T
as an R
-algebra.
Equations
Instances For
If S
is the localization of R
at r
, this is the canonical submersive presentation
of S
as R
-algebra.
Equations
- Algebra.SubmersivePresentation.localizationAway S r = { toPreSubmersivePresentation := Algebra.PreSubmersivePresentation.localizationAway S r, jacobian_isUnit := ⋯ }
Instances For
If P
is a submersive presentation of S
over R
and T
is an R
-algebra, we
obtain a natural submersive presentation of T ⊗[R] S
over T
.
Equations
- Algebra.SubmersivePresentation.baseChange T P = { toPreSubmersivePresentation := Algebra.PreSubmersivePresentation.baseChange T P.toPreSubmersivePresentation, jacobian_isUnit := ⋯ }
Instances For
Given a submersive presentation P
and equivalences ι ≃ P.vars
and
κ ≃ σ
, this is the induced sumbersive presentation with variables indexed
by ι
and relations indexed by `κ
Instances For
If P
is submersive, PreSubmersivePresentation.aevalDifferential
is an isomorphism.
Equations
Instances For
If P
is a submersive presentation, the partial derivatives of P.relation i
by
P.map j
form a basis of σ → S
.
Equations
- P.basisDeriv = (Pi.basisFun S σ).map P.aevalDifferentialEquiv
Instances For
The relative dimension of a standard smooth R
-algebra S
is
the dimension of an arbitrarily chosen submersive R
-presentation of S
.
Note: If S
is non-trivial, this number is independent of the choice of the presentation as it is
equal to the S
-rank of Ω[S/R]
(see IsStandardSmoothOfRelativeDimension.rank_kaehlerDifferential
).