Order ideals of vector lattices #
An order ideal (or simply ideal) of a vector lattice is a sublattice that is solid:
if x ∈ J and |y| ≤ |x| then y ∈ J. Equivalently, an ideal is precisely a solid
subspace. This file defines the bundled OrderIdeal structure extending
VectorSublattice and establishes the basic characterisations and properties.
An OrderIdeal of a vector lattice X is a vector sublattice that is solid:
whenever x ∈ J and 0 ≤ y ≤ x, we have y ∈ J.
Instances For
Equations
- OrderIdeal.instSetLike = { coe := fun (J : OrderIdeal X) => J.carrier, coe_injective := ⋯ }
Order ideals of X, ordered by inclusion, form a partial order.
Equations
An order ideal is closed under ⊔.
An order ideal is closed under ⊓.
An order ideal is closed under absolute value.
Solidity in terms of absolute value: x ∈ J and |y| ≤ |x| imply
y ∈ J.
Construction from solidity #
Build an OrderIdeal from a submodule that is solid in the absolute-value
sense: x ∈ M and |y| ≤ |x| imply y ∈ M. Every solid subspace is
automatically a sublattice and an ideal.
Equations
- OrderIdeal.ofSolid M h = { toSubmodule := M, sup_mem' := ⋯, solid' := ⋯ }
Instances For
A submodule is the carrier of an order ideal iff it is solid.
Ideal generated by a set #
The intersection of two order ideals is an order ideal.
Equations
- J₁.inf J₂ = { toSubmodule := J₁.toSubmodule ⊓ J₂.toSubmodule, sup_mem' := ⋯, solid' := ⋯ }
Instances For
Order ideals of X admit arbitrary intersections: they form a complete
semilattice for the ⊓ operation.
Equations
- One or more equations did not get rendered due to their size.
Membership in an arbitrary intersection of order ideals.
The ideal generated by a set s ⊆ X is the smallest order ideal of
X containing s, defined as the intersection of all order ideals containing
s.
Equations
- OrderIdeal.generated s = sInf {J : OrderIdeal X | s ⊆ ↑J}
Instances For
The set s is contained in the ideal it generates.
The ideal generated by s is contained in any ideal containing s.
Explicit description of the generated ideal. An element x lies in the
ideal generated by s if and only if |x| is bounded above by a non-negative
linear combination of absolute values of elements of s.
Principal ideal #
The principal ideal generated by a is the set of elements x with
|x| ≤ c • |a| for some c ≥ 0.
Equations
Instances For
Characterisation of membership in the principal ideal.
The generator belongs to its own principal ideal.
The principal ideal generated by a coincides with the ideal generated by
the singleton {|a|}.
Gauge norm on the principal ideal #
The gauge norm (or order-unit norm) of x with respect to a is
inf {c ≥ 0 | |x| ≤ c • |a|}. For x in the principal ideal of a, this
is finite and defines a lattice seminorm; it is a norm precisely when the
ambient space is Archimedean.
Instances For
The gauge norm is non-negative.
The gauge norm of zero is zero.
The gauge norm is symmetric.
The gauge norm controls the absolute value: |x| ≤ ‖x‖_a • |a| for
x in the principal ideal.
Any admissible constant bounds the gauge norm from above.
The gauge norm is monotone with respect to |·|: the solid-norm
property.
In an Archimedean vector lattice the gauge norm is definite:
gaugeNorm a x = 0 ↔ x = 0 for x in the principal ideal of a.
The closed unit ball of the gauge norm is the order interval
[-|a|, |a|].
If 0 ≤ e ≤ u then ‖·‖_u ≤ ‖·‖_e on I_e.
Normed vector lattice structure on the principal ideal #
The underlying submodule of the principal ideal.
Equations
Instances For
The gauge norm as a Norm instance on the principal ideal.
Equations
- OrderIdeal.instNormPrincipal a = { norm := fun (x : ↥(OrderIdeal.principalSubmodule a)) => OrderIdeal.gaugeNorm a ↑x }
The principal ideal inherits a lattice structure from X.
Equations
- One or more equations did not get rendered due to their size.
The principal ideal is an ordered additive monoid.
In an Archimedean vector lattice, the principal ideal I_a with the gauge
norm is a normed additive commutative group.
Instances For
In an Archimedean vector lattice, the principal ideal I_a with the gauge
norm admits a VectorLattice structure.
Equations
- OrderIdeal.principalVectorLattice a = { toModule := (OrderIdeal.principalSubmodule a).module, toPosSMulMono := ⋯ }
Instances For
In an Archimedean vector lattice, the principal ideal I_a equipped with
the gauge norm is a normed vector lattice.
Equations
- OrderIdeal.principalNormedVectorLattice a = { toVectorLattice := OrderIdeal.principalVectorLattice a, toHasSolidNorm := ⋯, toNormSMulClass := ⋯ }
Instances For
Sum of ideals #
The sum of two order ideals (as submodules) is again an order ideal.
Equations
- J₁.sum J₂ = OrderIdeal.ofSolid (J₁.toSubmodule + J₂.toSubmodule) ⋯
Instances For
The underlying submodule of sum J₁ J₂ is J₁.toSubmodule + J₂.toSubmodule.
Positive decomposition in the sum of two ideals: every non-negative
element of J₁ + J₂ admits a splitting y = y₁ + y₂ with 0 ≤ y₁ ∈ J₁ and
0 ≤ y₂ ∈ J₂.
Bounded decomposition in the sum of two ideals: every element of
J₁ + J₂ admits a splitting z = a + b with a ∈ J₁, b ∈ J₂ and the
lattice estimates |a| ≤ |z|, |b| ≤ |z|.
Complete lattice structure #
The whole space X is an order ideal.
The trivial ideal {0} is an order ideal.
Order ideals of X, ordered by inclusion, form a complete lattice.
Binary joins are given by the (Minkowski) sum sum, binary meets and arbitrary
infima are given by intersection, the bottom element is the trivial ideal
{0}, and the top element is the whole space.
Equations
- One or more equations did not get rendered due to their size.
Strong order units and the principal ideal #
An element is a strong order unit precisely when it is non-negative and the principal ideal it generates is the whole space.
Existence of proper non-trivial ideals #
If the real dimension of X is strictly greater than one, then X admits
an order ideal that is neither trivial nor the whole space.
Closure of an ideal in a normed vector lattice #
The topological closure of the underlying submodule of an order ideal is
itself solid: if x lies in the closure and |y| ≤ |x|, then y lies in the
closure.
The norm closure of an order ideal J in a normed vector lattice is
again an order ideal, whose underlying submodule is the topological closure of
J.toSubmodule.
Equations
Instances For
A closed order ideal of a Banach lattice is an order ideal whose underlying set is closed in the norm topology.
- isClosed' : IsClosed ↑self.toOrderIdeal
Instances For
Equations
- ClosedOrderIdeal.instSetLike = { coe := fun (J : ClosedOrderIdeal X) => ↑J.toOrderIdeal, coe_injective := ⋯ }
A closed order ideal is closed as a subset of X.
Closed order ideals form a partial order under inclusion.
The intersection of two closed order ideals is a closed order ideal.
Equations
- J₁.inf J₂ = { toOrderIdeal := J₁.toOrderIdeal ⊓ J₂.toOrderIdeal, isClosed' := ⋯ }
Instances For
The sum of two closed order ideals in a Banach lattice is again closed.
The sum of two closed order ideals of a Banach lattice is again a closed order ideal.
Instances For
The arbitrary intersection of a family of closed order ideals is a closed order ideal.
Equations
- ClosedOrderIdeal.sInf S = { toOrderIdeal := sInf ((fun (J : ClosedOrderIdeal X) => J.toOrderIdeal) '' S), isClosed' := ⋯ }
Instances For
Closed order ideals of a Banach lattice form a lattice under inclusion, with binary meets given by intersection and binary joins by the sum.
Equations
- One or more equations did not get rendered due to their size.
Closed order ideals of a Banach lattice admit arbitrary intersections:
they form a complete semilattice for the ⊓ operation.
Equations
Membership in an arbitrary intersection of closed order ideals.
The underlying submodule of a binary meet is the intersection of the underlying submodules.
The underlying submodule of a binary join is the sum of the underlying submodules.