Sublattices of vector lattices #
A vector sublattice of a vector lattice is a linear subspace closed under the lattice
operations. Since all lattice operations are expressible in terms of each other, it suffices
that the subspace be closed under any one operation — for example, ⊔ or |·|. The key
characterisation proved here is that closure under absolute value is equivalent to the
sublattice property. The file also constructs generated sublattices, describes
them in terms of sup- and inf-closures, and records the induced normed vector
lattice structure on closed sublattices.
Sup- and inf-closure of a pointed cone #
In a vector lattice, the sup-closure and inf-closure of a pointed cone
C ⊆ X are again pointed cones.
If s is closed under addition, so is its sup-closure.
If s is closed under non-negative scaling, so is its sup-closure.
If s is closed under addition, so is its inf-closure.
If s is closed under non-negative scaling, so is its inf-closure.
The sup-closure of a pointed cone is a pointed cone.
Equations
- C.supClosure = { carrier := supClosure ↑C, add_mem' := ⋯, zero_mem' := ⋯, smul_mem' := ⋯ }
Instances For
The inf-closure of a pointed cone is a pointed cone.
Equations
- C.infClosure = { carrier := infClosure ↑C, add_mem' := ⋯, zero_mem' := ⋯, smul_mem' := ⋯ }
Instances For
A VectorSublattice of a vector lattice X is a linear subspace closed under
⊔. This is the bundled version: it extends Submodule ℝ X with lattice closure.
Instances For
Equations
- VectorSublattice.instSetLike = { coe := fun (Y : VectorSublattice X) => Y.carrier, coe_injective := ⋯ }
A vector sublattice is closed under ⊔.
A vector sublattice is closed under ⊓.
A vector sublattice is closed under the positive part.
A vector sublattice is closed under the negative part.
A vector sublattice is closed under absolute value.
The underlying set of a vector sublattice is an IsSublattice.
Construction from absolute-value closure #
Build a VectorSublattice from a submodule closed under |·|.
Equations
- VectorSublattice.ofAbsClosed M h = { toSubmodule := M, sup_mem' := ⋯ }
Instances For
A submodule is a vector sublattice iff it is closed under absolute value.
Coercion to submodule #
A vector sublattice and its underlying submodule have the same carrier.
Coercion to pointed cone #
Every vector sublattice, being a linear subspace, is in particular a pointed convex cone.
Every vector sublattice is (canonically) a pointed cone.
Equations
- VectorSublattice.instCoeHeadPointedConeReal = { coe := fun (Y : VectorSublattice X) => let __AddSubmonoid := Y.toAddSubmonoid; { toAddSubmonoid := __AddSubmonoid, smul_mem' := ⋯ } }
Lattice structure on the underlying subtype #
A vector sublattice Y inherits a lattice and vector-lattice structure from
the ambient space, with ⊔ and ⊓ computed pointwise.
The lattice structure on the underlying subtype of a vector sublattice.
Equations
The subtype of a vector sublattice is an ordered additive monoid.
Scalar multiplication by non-negative reals is monotone on the subtype of a vector sublattice.
The subtype of a vector sublattice is itself a vector lattice.
Equations
- Y.instVectorLatticeSubtype = { toModule := Y.module, toPosSMulMono := ⋯ }
Lattice structure on VectorSublattice X #
The collection of vector sublattices of X is ordered by inclusion. It
contains the whole space ⊤ = X and the zero subspace ⊥ = {0}, and is
closed under arbitrary intersections.
Equations
- VectorSublattice.instPartialOrder = PartialOrder.lift (fun (Y : VectorSublattice X) => ↑Y) ⋯
The whole space X is a vector sublattice.
Every element belongs to ⊤.
The zero subspace {0} is a vector sublattice.
An element of ⊥ is zero.
Arbitrary intersections of vector sublattices are vector sublattices.
Equations
- One or more equations did not get rendered due to their size.
The vector sublattice generated by a set #
The vector sublattice generated by a set s ⊆ X is by definition the
smallest vector sublattice of X containing s, obtained as the infimum
of all vector sublattices containing s.
The vector sublattice generated by a set s: the smallest vector
sublattice of X containing s.
Equations
- VectorSublattice.generated s = sInf {Y : VectorSublattice X | s ⊆ ↑Y}
Instances For
The generating set is contained in the vector sublattice it generates.
The vector sublattice generated by s is contained in every vector
sublattice containing s.
The vector sublattice generated by a linear subspace M coincides with
the sup-closure of the inf-closure of M.
The vector sublattice generated by a linear subspace M coincides with
the inf-closure of the sup-closure of M.
The vector sublattice generated by a pointed cone C coincides with the
difference set of the sup-closure of C with itself.
The vector sublattice generated by a pointed cone C coincides with the
difference set of the inf-closure of C with itself.
The vector sublattice generated by a set of pairwise lattice-disjoint, non-negative elements coincides with its linear span.
The sublattice generated by the image of a tuple x : Fin n → X coincides
with the set of lattice-linear combinations of x.
The sublattice generated by a set s coincides with the union, over all
natural numbers n and all n-tuples x with values in s, of the sets of
lattice-linear combinations of x.
Normed and Banach lattice structure on sublattices #
A vector sublattice of a normed vector lattice is itself a normed vector lattice under the induced norm and lattice operations.
Equations
- Y.instNormedVectorLatticeSubtype = { toVectorLattice := Y.instVectorLatticeSubtype, toHasSolidNorm := ⋯, toNormSMulClass := ⋯ }
The closure of a vector sublattice in a normed vector lattice is again a vector sublattice.
Equations
- Y.topologicalClosure = { toSubmodule := Y.topologicalClosure, sup_mem' := ⋯ }
Instances For
A norm-closed vector sublattice of a Banach lattice is itself a Banach lattice under the induced structures.
Equations
- Y.banachLatticeSubtype hclosed = { toNormedVectorLattice := Y.instNormedVectorLatticeSubtype, toCompleteSpace := ⋯ }
Instances For
The closed vector sublattice generated by a countable subset of a normed vector lattice is separable.