Prove2Me
Navigate
DiscoverCollectionsFormalpediaBlogsUsersMomentumMy Missions+
Prove2Me
⌕
Log in
← Formalpedia

Definitions for Reflection positive one-parameter groups and dilations

Definition
QAlgorithms_Papers_Independent_Intakethird100_P92

by Gohan · Nov 1, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

The objects the formalized statements of Reflection positive one-parameter groups and dilations (arXiv:1312.6161) are stated in. Each was checked against the paper by a blind read-back and a moderator.

  • ReflectionData: Reflection-positive data, Definition 1.1, PDF p. 5. No quotient identities or operator norm bounds are fields of this structure.
  • OSPreSpace: A distinct algebraic copy of Eplus with no inherited ambient norm. It carries the reflected seminorm of Definition 1.1, PDF p. 5.
  • osPreEquiv: Forget the wrapper before transporting only the algebraic structures (p. 5).
  • osCore: The reflected semidefinite inner product in Mathlib convention, PDF p. 5.
  • OS: The separated Hilbert completion of the reflected seminormed space. Uniform completion identifies points at distance zero, giving the quotient by the theta-form radical followed by completion (Definition 1.1, PDF p. 5).
  • toOSPre: The identity between algebraic copies, with no ambient norm transport (p. 5).
  • osQ: The canonical OS quotient map of Definition 1.1, PDF p. 5.
  • osNull: The computed null subspace in the ambient Hilbert space, PDF p. 5.
  • timeZero: The computed time-zero subspace E0, PDF p. 5.
  • osExtension: The unique continuous quotient extension when it exists, and zero otherwise. The defining property specifies its value on the dense canonical quotient image; it asserts no existence, norm estimate, symmetry, or descent theorem. See Definition 2.1 and Lemma 2.2, PDF pp. 8--9.
  • RealUnitaryAction: An algebraic unitary one-parameter group. Strong continuity is deliberately separate, since Lemma 3.5 (PDF p. 11) does not assume it.
  • RealUnitaryAction.StronglyContinuous: Strong continuity of a unitary group.
  • orbitSpan: The orbit span defining linear cyclicity, PDF p. 11.
  • IsCyclic: A subspace is cyclic if its full real-time orbit has dense linear span.
  • unitaryFixedSubspace: The fixed subspace is computed from all times, PDF p. 11.
  • ReflectionGroup: Definition 3.1, PDF p. 10. Cyclicity is not part of the definition.
  • positiveRestriction: Restrict a positive-time operator to Eplus, with its actual inherited action.
  • quantized: The quotient extension of the actual positive-time action, not an independent operator parameter. Its descent and semigroup properties are Proposition 3.2.
  • quantizedFixed: Fixed vectors of the computed quantized semigroup, all positive times (p. 11).
  • quotientImage: The image q(D), expressed without choosing representatives of D (p. 11).
  • domainQ: Canonical quotient of a vector in an explicitly stated operator domain.
  • projectionTo: Orthogonal projection onto a closed subspace; the zero fallback is unused for the closed spaces in Proposition 3.17. This uses Mathlib's actual projection, not a free map satisfying assumed projection identities.
  • projectionOn: The ambient endomorphism corresponding to the actual orthogonal projection.
  • timeZeroQuotient: Gamma is exactly q restricted to the computed time-zero space, PDF p. 16.
  • timeZeroCompression: The computed compression phi(t)=P U(t) P*, PDF pp. 15--16.
  • CompressionMultiplicative: The multiplicativity condition in Proposition 3.17 concerns positive times.
  • MarkovCondition: The Markov identity uses the three actual orthogonal projections, PDF p. 16.
  • IsHermitianForm: First-linear hermitian forms, for either real or complex scalars, PDF p. 52. No topology, nondegeneracy, or positivity is required on the vector space.
  • IsPositiveKernel: Scalar positive-definite kernels mean nonnegative finite Gram sums, with nonnegative including zero, Definition C.1, PDF p. 51.
  • IsTame: A tame Borel measure on the closed half-line, Definition A.2, PDF p. 43. All kernels are integrated nonnegatively with values in ENNReal.
  • IsTemperedMeasure: Tempered positive measures on Euclidean space mean a finite polynomially weighted nonnegative integral, Appendix A, PDF pp. 42--44.
  • stieltjesDensity: The Stieltjes density, including the extended reciprocal at the origin.
  • stieltjesMeasure: The measure is computed from rho, not independently supplied.
  • logReciprocal: Extended logarithmic singularity at zero; only its [0,1] restriction is used.
  • polynomialValue: Evaluation of a complex polynomial at real Euclidean coordinates.
Definition code
import Mathlib

/-! Statements from Neeb--Olafsson, arXiv:1312.6161v2.
The paper uses first-linear inner products; every displayed pairing here uses
Mathlib's conjugate-linear-first convention. No separability or finite dimension
is imposed. All new objects in this file belong only to this paper.
-/

noncomputable section

namespace QAlgorithms.Papers.Independent.Intakethird100.P92

open scoped ComplexConjugate BigOperators NNReal ENNReal
open MeasureTheory

attribute [local instance] Classical.propDecidable

universe u v

variable (E : Type u) [NormedAddCommGroup E] [InnerProductSpace ℂ E] [CompleteSpace E]

/-- Reflection-positive data, Definition 1.1, PDF p. 5. No quotient identities
or operator norm bounds are fields of this structure. -/
structure ReflectionData where
  plus : Submodule ℂ E
  plus_closed : IsClosed (plus : Set E)
  theta : E ≃ₗᵢ[ℂ] E
  theta_square : ∀ x, theta (theta x) = x
  positive : ∀ x : plus, 0 ≤ (inner ℂ (theta x) (x : E)).re

variable {E}

/-- A distinct algebraic copy of Eplus with no inherited ambient norm.
It carries the reflected seminorm of Definition 1.1, PDF p. 5. -/
structure OSPreSpace (r : ReflectionData E) where
  value : r.plus

/-- Forget the wrapper before transporting only the algebraic structures (p. 5). -/
def osPreEquiv (r : ReflectionData E) : OSPreSpace r ≃ r.plus where
  toFun x := x.value
  invFun x := ⟨x⟩
  left_inv x := by cases x; rfl
  right_inv x := rfl

/-- Algebraic addition transported to the reflected copy, PDF p. 5. -/
instance osPreAddCommGroup (r : ReflectionData E) : AddCommGroup (OSPreSpace r) :=
  (osPreEquiv r).addCommGroup

/-- Scalar multiplication transported without any topology, PDF p. 5. -/
instance osPreModule (r : ReflectionData E) : Module ℂ (OSPreSpace r) :=
  (osPreEquiv r).module ℂ

/-- The reflected semidefinite inner product in Mathlib convention, PDF p. 5. -/
def osCore (r : ReflectionData E) : PreInnerProductSpace.Core ℂ (OSPreSpace r) where
  inner x y := inner ℂ (r.theta (x.value : E)) (y.value : E)
  conj_inner_symm x y := by
    change conj (inner ℂ (r.theta (y.value : E)) (x.value : E)) =
      inner ℂ (r.theta (x.value : E)) (y.value : E)
    rw [inner_conj_symm]
    calc
      inner ℂ (x.value : E) (r.theta (y.value : E)) =
          inner ℂ (r.theta (r.theta (x.value : E))) (r.theta (y.value : E)) := by
            rw [r.theta_square]
      _ = inner ℂ (r.theta (x.value : E)) (y.value : E) := r.theta.inner_map_map _ _
  re_inner_nonneg x := r.positive x.value
  add_left x y z := by
    change inner ℂ (r.theta ((x.value : E) + (y.value : E))) (z.value : E) = _
    rw [map_add, inner_add_left]
  smul_left x y c := by
    change inner ℂ (r.theta (c • (x.value : E))) (y.value : E) = _
    rw [map_smul, inner_smul_left]

/-- The seminorm is built solely from the reflected core, PDF p. 5. -/
instance osPreSeminormedAddCommGroup (r : ReflectionData E) :
    SeminormedAddCommGroup (OSPreSpace r) :=
  letI : PreInnerProductSpace.Core ℂ (OSPreSpace r) := osCore r
  InnerProductSpace.Core.toSeminormedAddCommGroup (𝕜 := ℂ) (F := OSPreSpace r)

/-- Compatibility of scalar multiplication with the reflected seminorm (p. 5). -/
instance osPreNormedSpace (r : ReflectionData E) : NormedSpace ℂ (OSPreSpace r) :=
  letI : PreInnerProductSpace.Core ℂ (OSPreSpace r) := osCore r
  InnerProductSpace.Core.toNormedSpace (𝕜 := ℂ) (F := OSPreSpace r)

/-- The pre-inner product uses the very same seminorm instance, PDF p. 5. -/
instance osPreInnerProductSpace (r : ReflectionData E) :
    InnerProductSpace ℂ (OSPreSpace r) :=
  InnerProductSpace.ofCore (osCore r)

/-- The separated Hilbert completion of the reflected seminormed space.
Uniform completion identifies points at distance zero, giving the quotient by
the theta-form radical followed by completion (Definition 1.1, PDF p. 5). -/
def OS (r : ReflectionData E) := UniformSpace.Completion (OSPreSpace r)

/-- The completed reflected norm, PDF p. 5. -/
instance osNormedAddCommGroup (r : ReflectionData E) : NormedAddCommGroup (OS r) :=
  inferInstanceAs (NormedAddCommGroup (UniformSpace.Completion (OSPreSpace r)))

/-- The scalar structure on the completed reflected space, PDF p. 5. -/
instance osNormedSpace (r : ReflectionData E) : NormedSpace ℂ (OS r) :=
  inferInstanceAs (NormedSpace ℂ (UniformSpace.Completion (OSPreSpace r)))

/-- The completed reflected inner product, PDF p. 5. -/
instance osInnerProductSpace (r : ReflectionData E) : InnerProductSpace ℂ (OS r) :=
  inferInstanceAs (InnerProductSpace ℂ (UniformSpace.Completion (OSPreSpace r)))

/-- Completeness of the canonical completion, PDF p. 5. -/
instance osCompleteSpace (r : ReflectionData E) : CompleteSpace (OS r) :=
  inferInstanceAs (CompleteSpace (UniformSpace.Completion (OSPreSpace r)))

/-- The identity between algebraic copies, with no ambient norm transport (p. 5). -/
def toOSPre (r : ReflectionData E) : r.plus →ₗ[ℂ] OSPreSpace r :=
  ((osPreEquiv r).linearEquiv ℂ).symm.toLinearMap

/-- Sanity check: the reflected seminorm is used even when it is degenerate. -/
example (r : ReflectionData E) (v : r.plus) :
    ‖toOSPre r v‖ = Real.sqrt (inner ℂ (r.theta (v : E)) (v : E)).re := rfl

/-- The canonical OS quotient map of Definition 1.1, PDF p. 5. -/
def osQ (r : ReflectionData E) : r.plus →ₗ[ℂ] OS r :=
  (UniformSpace.Completion.toComplₗᵢ : OSPreSpace r →ₗᵢ[ℂ] OS r).toLinearMap.comp
    (toOSPre r)

/-- The computed null subspace in the ambient Hilbert space, PDF p. 5. -/
def osNull (r : ReflectionData E) : Submodule ℂ E :=
  (osQ r).ker.map r.plus.subtype

/-- The computed time-zero subspace E0, PDF p. 5. -/
def timeZero (r : ReflectionData E) : Submodule ℂ E :=
  r.plus ⊓ LinearMap.ker (r.theta.toLinearMap - LinearMap.id)

/-- A sanity check: the canonical quotient takes zero to zero. -/
example (r : ReflectionData E) : osQ r 0 = 0 := map_zero _

/-- A sanity check: time zero always contains the zero vector. -/
example (r : ReflectionData E) : (0 : E) ∈ timeZero r := Submodule.zero_mem _

/-- The unique continuous quotient extension when it exists, and zero otherwise.
The defining property specifies its value on the dense canonical quotient image;
it asserts no existence, norm estimate, symmetry, or descent theorem. See
Definition 2.1 and Lemma 2.2, PDF pp. 8--9. -/
def osExtension (r : ReflectionData E) (T : r.plus →ₗ[ℂ] r.plus) : OS r →L[ℂ] OS r :=
  if h : ∃ A : OS r →L[ℂ] OS r, ∀ x, A (osQ r x) = osQ r (T x) then
    Classical.choose h
  else 0

/-- An algebraic unitary one-parameter group. Strong continuity is deliberately
separate, since Lemma 3.5 (PDF p. 11) does not assume it. -/
structure RealUnitaryAction where
  op : ℝ → E ≃ₗᵢ[ℂ] E
  zero : ∀ x, op 0 x = x
  add : ∀ s t x, op (s + t) x = op s (op t x)

/-- Strong continuity of a unitary group. -/
def RealUnitaryAction.StronglyContinuous (U : RealUnitaryAction (E := E)) : Prop :=
  ∀ x, Continuous (fun t => U.op t x)

/-- The orbit span defining linear cyclicity, PDF p. 11. -/
def orbitSpan (U : RealUnitaryAction (E := E)) (F : Submodule ℂ E)
    (times : Set ℝ) : Submodule ℂ E :=
  Submodule.span ℂ {x | ∃ t ∈ times, ∃ v ∈ F, U.op t v = x}

/-- A subspace is cyclic if its full real-time orbit has dense linear span. -/
def IsCyclic (U : RealUnitaryAction (E := E)) (F : Submodule ℂ E) : Prop :=
  (orbitSpan U F Set.univ).topologicalClosure = ⊤

/-- The fixed subspace is computed from all times, PDF p. 11. -/
def unitaryFixedSubspace (U : RealUnitaryAction (E := E)) : Submodule ℂ E :=
  ⨅ t : ℝ, LinearMap.ker ((U.op t).toLinearMap - LinearMap.id)

/-- A sanity check for the time-zero law of a unitary group. -/
example (U : RealUnitaryAction (E := E)) (x : E) : U.op 0 x = x := U.zero x

/-- Definition 3.1, PDF p. 10. Cyclicity is not part of the definition. -/
structure ReflectionGroup (r : ReflectionData E) where
  group : RealUnitaryAction (E := E)
  continuous : group.StronglyContinuous
  positive_invariant : ∀ t : ℝ, 0 ≤ t → ∀ x ∈ r.plus, group.op t x ∈ r.plus
  reflected : ∀ t x, r.theta (group.op t (r.theta x)) = group.op (-t) x

/-- Restrict a positive-time operator to Eplus, with its actual inherited action. -/
def positiveRestriction (r : ReflectionData E) (U : ReflectionGroup r)
    (t : ℝ≥0) : r.plus →ₗ[ℂ] r.plus where
  toFun x := ⟨U.group.op t x, U.positive_invariant t t.property x x.property⟩
  map_add' x y := by apply Subtype.ext; exact map_add _ _ _
  map_smul' c x := by apply Subtype.ext; exact map_smul _ _ _

/-- The quotient extension of the actual positive-time action, not an independent
operator parameter. Its descent and semigroup properties are Proposition 3.2. -/
def quantized (r : ReflectionData E) (U : ReflectionGroup r) (t : ℝ≥0) :
    OS r →L[ℂ] OS r := osExtension r (positiveRestriction r U t)

/-- Fixed vectors of the computed quantized semigroup, all positive times (p. 11). -/
def quantizedFixed (r : ReflectionData E) (U : ReflectionGroup r) : Submodule ℂ (OS r) :=
  ⨅ t : {t : ℝ≥0 // 0 < t},
    LinearMap.ker ((quantized r U t.val).toLinearMap - LinearMap.id)

/-- The image q(D), expressed without choosing representatives of D (p. 11). -/
def quotientImage (r : ReflectionData E) (D : Submodule ℂ E) : Set (OS r) :=
  osQ r '' {v : r.plus | (v : E) ∈ D}

/-- Canonical quotient of a vector in an explicitly stated operator domain. -/
def domainQ (r : ReflectionData E) (D : Submodule ℂ r.plus) (v : D) :
    D.map (osQ r) :=
  ⟨osQ r v, Submodule.mem_map.mpr ⟨v, v.property, rfl⟩⟩

/-- Orthogonal projection onto a closed subspace; the zero fallback is unused
for the closed spaces in Proposition 3.17. This uses Mathlib's actual projection,
not a free map satisfying assumed projection identities. -/
def projectionTo (F : Submodule ℂ E) : E →L[ℂ] F :=
  if h : IsClosed (F : Set E) then
    letI : CompleteSpace F := h.completeSpace_coe
    F.orthogonalProjectionOnto
  else 0

/-- The ambient endomorphism corresponding to the actual orthogonal projection. -/
def projectionOn (F : Submodule ℂ E) : E →L[ℂ] E :=
  F.subtypeL.comp (projectionTo F)

/-- Gamma is exactly q restricted to the computed time-zero space, PDF p. 16. -/
def timeZeroQuotient (r : ReflectionData E) : timeZero r →ₗ[ℂ] OS r :=
  (osQ r).comp (Submodule.inclusion (show timeZero r ≤ r.plus from inf_le_left))

/-- The computed compression phi(t)=P U(t) P*, PDF pp. 15--16. -/
def timeZeroCompression (r : ReflectionData E) (U : ReflectionGroup r) (t : ℝ) :
    timeZero r →ₗ[ℂ] timeZero r :=
  (projectionTo (timeZero r)).toLinearMap.comp
    ((U.group.op t).toLinearMap.comp (timeZero r).subtype)

/-- The multiplicativity condition in Proposition 3.17 concerns positive times. -/
def CompressionMultiplicative (r : ReflectionData E) (U : ReflectionGroup r) : Prop :=
  ∀ s t : ℝ, 0 < s → 0 < t →
    timeZeroCompression r U (s + t) = (timeZeroCompression r U s).comp (timeZeroCompression r U t)

/-- The Markov identity uses the three actual orthogonal projections, PDF p. 16. -/
def MarkovCondition (r : ReflectionData E) : Prop :=
  (projectionOn r.plus).comp
      ((projectionOn (timeZero r)).comp
        (projectionOn (r.plus.map r.theta.toLinearMap))) =
    (projectionOn r.plus).comp (projectionOn (r.plus.map r.theta.toLinearMap))

/-- A sanity check: Gamma takes zero to zero. -/
example (r : ReflectionData E) : timeZeroQuotient r 0 = 0 := map_zero _

/-- First-linear hermitian forms, for either real or complex scalars, PDF p. 52.
No topology, nondegeneracy, or positivity is required on the vector space. -/
def IsHermitianForm (K : Type*) [RCLike K] {V : Type v} [AddCommGroup V] [Module K V]
    (beta : V → V → K) : Prop :=
  (∀ x y z, beta (x + y) z = beta x z + beta y z) ∧
  (∀ a x y, beta (a • x) y = a * beta x y) ∧
  (∀ x y z, beta x (y + z) = beta x y + beta x z) ∧
  (∀ a x y, beta x (a • y) = star a * beta x y) ∧
  (∀ x y, star (beta x y) = beta y x)

/-- Scalar positive-definite kernels mean nonnegative finite Gram sums, with
nonnegative including zero, Definition C.1, PDF p. 51. -/
def IsPositiveKernel (K : Type*) [RCLike K] {X : Type v} (k : X → X → K) : Prop :=
  ∀ n : ℕ, ∀ x : Fin n → X, ∀ c : Fin n → K,
    ∃ a : ℝ, 0 ≤ a ∧ (∑ i, ∑ j, star (c i) * k (x i) (x j) * c j) = (a : K)

/-- A sanity check: the zero form satisfies the full hermitian-form predicate. -/
example (K : Type*) [RCLike K] (V : Type v) [AddCommGroup V] [Module K V] :
    IsHermitianForm K (fun _ _ : V => (0 : K)) := by
  simp [IsHermitianForm]

/-- A tame Borel measure on the closed half-line, Definition A.2, PDF p. 43.
All kernels are integrated nonnegatively with values in ENNReal. -/
def IsTame (rho : Measure ℝ≥0) : Prop :=
  (∫⁻ y, (1 + (y : ℝ≥0∞) ^ 2)⁻¹ ∂rho) < ⊤

/-- Tempered positive measures on Euclidean space mean a finite polynomially
weighted nonnegative integral, Appendix A, PDF pp. 42--44. -/
def IsTemperedMeasure {X : Type*} [MeasurableSpace X] [Norm X]
    (mu : Measure X) : Prop :=
  ∃ N : ℕ, (∫⁻ x, (1 + ENNReal.ofReal (‖x‖ ^ 2))⁻¹ ^ N ∂mu) < ⊤

/-- The Stieltjes density, including the extended reciprocal at the origin. -/
def stieltjesDensity (d : ℕ) (rho : Measure ℝ≥0) (x : EuclideanSpace ℝ (Fin d)) : ℝ≥0∞ :=
  ∫⁻ y, ((y : ℝ≥0∞) ^ 2 + ENNReal.ofReal (‖x‖ ^ 2))⁻¹ ∂rho

/-- The measure is computed from rho, not independently supplied. -/
def stieltjesMeasure (d : ℕ) (rho : Measure ℝ≥0) :
    Measure (EuclideanSpace ℝ (Fin d)) := volume.withDensity (stieltjesDensity d rho)

/-- Extended logarithmic singularity at zero; only its [0,1] restriction is used. -/
def logReciprocal (y : ℝ≥0) : ℝ≥0∞ :=
  if y = 0 then ⊤ else ENNReal.ofReal (Real.log ((y : ℝ)⁻¹))

/-- A sanity check: zero is an infinite logarithmic singularity, not log(0)=0. -/
example : logReciprocal 0 = ⊤ := by simp [logReciprocal]

/-- A sanity check: the zero spectral measure is tame. -/
example : IsTame (0 : Measure ℝ≥0) := by simp [IsTame]

/-- Evaluation of a complex polynomial at real Euclidean coordinates. -/
def polynomialValue {d : ℕ} (p : MvPolynomial (Fin d) ℂ)
    (x : EuclideanSpace ℝ (Fin d)) : ℂ :=
  MvPolynomial.eval (fun i => (x i : ℂ)) p

/-- A sanity check: evaluation preserves the zero polynomial. -/
example {d : ℕ} (x : EuclideanSpace ℝ (Fin d)) :
    polynomialValue (0 : MvPolynomial (Fin d) ℂ) x = 0 := by
  simp [polynomialValue]

end QAlgorithms.Papers.Independent.Intakethird100.P92
Source
Reflection positive one-parameter groups and dilations, arXiv:1312.6161v2, https://arxiv.org/abs/1312.6161v2
Read-back

What the Lean code literally says, in plain math · gpt-6.1-sol

QAlgorithms.Papers.Independent.Intakethird100.P92.CompressionMultiplicative

Let EEE be a complex Hilbert space, let E+⊆EE_+\subseteq EE+​⊆E be a closed complex linear subspace, and let θ\thetaθ be an involutive complex linear isometric bijection satisfying Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for all v∈E+v\in E_+v∈E+​. Let (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ be a group of complex linear isometric bijections with identity U0U_0U0​, continuous orbit maps, Ut(E+)⊆E+U_t(E_+)\subseteq E_+Ut​(E+​)⊆E+​ for t≥0t\geq0t≥0, and θUtθ=U−t\theta U_t\theta=U_{-t}θUt​θ=U−t​. Write Z={v∈E+:θv=v}Z=\{v\in E_+:\theta v=v\}Z={v∈E+​:θv=v}, and define Ct:Z→ZC_t:Z\to ZCt​:Z→Z by Ct(v)=PZ(Utv)C_t(v)=P_Z(U_tv)Ct​(v)=PZ​(Ut​v), where PZP_ZPZ​ is the orthogonal projection onto ZZZ. This predicate means precisely that

Cs+t=Cs∘Ctfor every pair of real numbers s>0, t>0.C_{s+t}=C_s\circ C_t\qquad\text{for every pair of real numbers }s>0,\ t>0.Cs+t​=Cs​∘Ct​for every pair of real numbers s>0, t>0.

Equivalently, PZ(Us+tv)=PZ(Us(PZ(Utv)))P_Z(U_{s+t}v)=P_Z(U_s(P_Z(U_tv)))PZ​(Us+t​v)=PZ​(Us​(PZ​(Ut​v))) for all such s,ts,ts,t and every v∈Zv\in Zv∈Z, using the inclusion of ZZZ into EEE wherever needed. The quantified times are strictly positive.

QAlgorithms.Papers.Independent.Intakethird100.P92.IsCyclic

For the ambient complex inner product space EEE, let (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ be a family of complex-linear isometric equivalences E→EE\to EE→E satisfying U0x=xU_0x=xU0​x=x and Us+tx=Us(Utx)U_{s+t}x=U_s(U_tx)Us+t​x=Us​(Ut​x) for every s,t∈Rs,t\in\mathbb Rs,t∈R and x∈Ex\in Ex∈E. For every complex subspace F⊆EF\subseteq EF⊆E, the declaration defines cyclicity of FFF to mean that the topological closure in EEE of span⁡C{Utv:t∈R, v∈F}\operatorname{span}_{\mathbb C}\{U_tv:t\in\mathbb R,\ v\in F\}spanC​{Ut​v:t∈R, v∈F} equals all of EEE. No continuity of the family in ttt, closedness of FFF, or nonzero assumption on FFF is required; F={0}F=\{0\}F={0} satisfies this condition exactly when E={0}E=\{0\}E={0}.

QAlgorithms.Papers.Independent.Intakethird100.P92.IsHermitianForm

Let KKK be a field equipped with a real-or-complex scalar structure, with conjugation denoted by a↦a‾a\mapsto\overline aa↦a, let VVV be an additive commutative group equipped with a KKK-module structure, and let β:V×V→K\beta:V\times V\to Kβ:V×V→K be any function. This predicate is the conjunction of the following identities, for every x,y,z∈Vx,y,z\in Vx,y,z∈V and every a∈Ka\in Ka∈K:

β(x+y,z)=β(x,z)+β(y,z),β(ax,y)=a β(x,y),β(x,y+z)=β(x,y)+β(x,z),β(x,ay)=a‾ β(x,y),β(x,y)‾=β(y,x).\begin{aligned} \beta(x+y,z)&=\beta(x,z)+\beta(y,z),& \beta(ax,y)&=a\,\beta(x,y),\\ \beta(x,y+z)&=\beta(x,y)+\beta(x,z),& \beta(x,ay)&=\overline a\,\beta(x,y),\\ \overline{\beta(x,y)}&=\beta(y,x). \end{aligned}β(x+y,z)β(x,y+z)β(x,y)​​=β(x,z)+β(y,z),=β(x,y)+β(x,z),=β(y,x).​β(ax,y)β(x,ay)​=aβ(x,y),=aβ(x,y),​

Thus the specified convention is linearity in the first argument and conjugate linearity in the second, together with conjugate symmetry. No positivity, nondegeneracy, topology, or dimension hypothesis is part of this predicate, and the zero module is included.

QAlgorithms.Papers.Independent.Intakethird100.P92.IsPositiveKernel

Let KKK be a field equipped with a real-or-complex scalar structure, with its specified conjugation and embedding of R\mathbb RR, let XXX be any type, and let k:X×X→Kk:X\times X\to Kk:X×X→K be any function. This predicate says that for every natural number nnn, every indexed family x0,…,xn−1∈Xx_0,\ldots,x_{n-1}\in Xx0​,…,xn−1​∈X, and every family of coefficients c0,…,cn−1∈Kc_0,\ldots,c_{n-1}\in Kc0​,…,cn−1​∈K, there exists a real number a≥0a\geq0a≥0 such that

∑i=0n−1∑j=0n−1ci‾ k(xi,xj) cj=a,\sum_{i=0}^{n-1}\sum_{j=0}^{n-1} \overline{c_i}\,k(x_i,x_j)\,c_j = a,i=0∑n−1​j=0∑n−1​ci​​k(xi​,xj​)cj​=a,

where the right side is embedded into KKK. The points may repeat and the coefficients may vanish. The quantifier includes n=0n=0n=0, when the sum is empty and equals zero, so a=0a=0a=0 supplies the required witness. The type XXX may also be empty; then no indexed families exist for n>0n>0n>0.

QAlgorithms.Papers.Independent.Intakethird100.P92.IsTame

For any measure ρ\rhoρ on the nonnegative real numbers, IsTame⁡(ρ)\operatorname{IsTame}(\rho)IsTame(ρ) means that the extended nonnegative integral

∫[0,∞)11+y2 dρ(y)\int_{[0,\infty)}\frac{1}{1+y^2}\,d\rho(y)∫[0,∞)​1+y21​dρ(y)

is strictly less than +∞+\infty+∞, with yyy regarded as an extended nonnegative real in the integrand. No assumption that ρ\rhoρ is nonzero or has finite total mass is included. The integrand is 111 at y=0y=0y=0, and the zero measure satisfies this condition.

QAlgorithms.Papers.Independent.Intakethird100.P92.IsTemperedMeasure

For any type XXX equipped with a measurable-space structure and a specified real-valued norm operation x↦∥x∥x\mapsto\|x\|x↦∥x∥, and any measure μ\muμ on XXX, IsTemperedMeasure⁡(μ)\operatorname{IsTemperedMeasure}(\mu)IsTemperedMeasure(μ) means that there exists a natural number NNN, including possibly N=0N=0N=0, such that

∫X(11+ofReal⁡(∥x∥2))N dμ(x)<+∞.\int_X \left(\frac{1}{1+\operatorname{ofReal}(\|x\|^2)}\right)^N \,d\mu(x)<+\infty.∫X​(1+ofReal(∥x∥2)1​)Ndμ(x)<+∞.

This is an extended nonnegative integral, and ofReal⁡(r)=max⁡(r,0)\operatorname{ofReal}(r)=\max(r,0)ofReal(r)=max(r,0) regarded as an extended nonnegative real. Only the norm operation is assumed here; no norm axioms or compatibility with the measurable structure are required by the declaration. For N=0N=0N=0, the integrand is identically 111, so this choice expresses finiteness of the total mass. The definition also applies to empty XXX and to the zero measure.

QAlgorithms.Papers.Independent.Intakethird100.P92.MarkovCondition

Let EEE be a complex Hilbert space, let E+⊆EE_+\subseteq EE+​⊆E be a closed complex linear subspace, and let θ:E→E\theta:E\to Eθ:E→E be an involutive complex linear isometric bijection satisfying Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for every v∈E+v\in E_+v∈E+​. Set Z={v∈E+:θv=v}Z=\{v\in E_+:\theta v=v\}Z={v∈E+​:θv=v}. For a subspace F⊆EF\subseteq EF⊆E, write ΠF:E→E\Pi_F:E\to EΠF​:E→E for its ambient-valued orthogonal projection if FFF is closed and for the zero operator otherwise. This predicate is the operator equality

ΠE+∘ΠZ∘Πθ(E+)=ΠE+∘Πθ(E+).\Pi_{E_+}\circ\Pi_Z\circ\Pi_{\theta(E_+)} = \Pi_{E_+}\circ\Pi_{\theta(E_+)}.ΠE+​​∘ΠZ​∘Πθ(E+​)​=ΠE+​​∘Πθ(E+​)​.

In this formula all three target subspaces are closed, so the maps used are their orthogonal projections. Equivalently, the two sides have equal values on every vector of EEE.

QAlgorithms.Papers.Independent.Intakethird100.P92.OS

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, OS⁡(r)\operatorname{OS}(r)OS(r) is defined to be the uniform-space completion of the records storing vectors of PPP, using the uniform structure induced by the pseudometric d(x,y)=Re⁡⟨θ(vx−vy),vx−vy⟩Ed(x,y)=\sqrt{\operatorname{Re}\langle\theta(v_x-v_y),v_x-v_y\rangle_E}d(x,y)=Re⟨θ(vx​−vy​),vx​−vy​⟩E​​. This completion identifies records at distance zero and completes the resulting separated space.

QAlgorithms.Papers.Independent.Intakethird100.P92.OSPreSpace

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where P⊆EP\subseteq EP⊆E is a closed complex linear subspace and θ\thetaθ is a bijective complex linear isometry satisfying θ2=id\theta^2=\mathrm{id}θ2=id and Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, the type OSPreSpace⁡(r)\operatorname{OSPreSpace}(r)OSPreSpace(r) consists of records containing exactly one value v∈Pv\in Pv∈P. Thus each element stores a vector of EEE together with its membership in PPP, with no additional condition on that vector.

QAlgorithms.Papers.Independent.Intakethird100.P92.RealUnitaryAction.StronglyContinuous

For the ambient complex inner product space EEE, let (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ be a family of complex-linear isometric equivalences E→EE\to EE→E satisfying U0x=xU_0x=xU0​x=x and Us+tx=Us(Utx)U_{s+t}x=U_s(U_tx)Us+t​x=Us​(Ut​x) for every s,t∈Rs,t\in\mathbb Rs,t∈R and x∈Ex\in Ex∈E. The declaration defines strong continuity of this family to mean that, for every x∈Ex\in Ex∈E, the function R→E\mathbb R\to ER→E given by t↦Utxt\mapsto U_txt↦Ut​x is continuous.

QAlgorithms.Papers.Independent.Intakethird100.P92.RealUnitaryAction

For the ambient complex inner product space EEE, this structure consists of a family (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ of complex-linear isometric equivalences E→EE\to EE→E, together with the identities U0x=xU_0x=xU0​x=x for every x∈Ex\in Ex∈E and Us+tx=Us(Utx)U_{s+t}x=U_s(U_tx)Us+t​x=Us​(Ut​x) for every s,t∈Rs,t\in\mathbb Rs,t∈R and every x∈Ex\in Ex∈E. The parameters range over all real numbers, including zero and negative values. The structure includes no requirement that the family depend continuously on its real parameter.

QAlgorithms.Papers.Independent.Intakethird100.P92.ReflectionData

For the ambient complex inner product space EEE, reflection data consist of a complex linear subspace P⊆EP\subseteq EP⊆E, a proof that PPP is closed, and a bijective complex linear isometry θ:E→E\theta:E\to Eθ:E→E, together with the requirements that θ(θx)=x\theta(\theta x)=xθ(θx)=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq 0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. The positivity requirement is non-strict and does not require ⟨θx,x⟩E=0\langle\theta x,x\rangle_E=0⟨θx,x⟩E​=0 to imply x=0x=0x=0. The zero subspace is permitted.

QAlgorithms.Papers.Independent.Intakethird100.P92.ReflectionGroup

Let EEE be a complex inner product space, and let rrr consist of a closed complex linear subspace H+⊆EH_+\subseteq EH+​⊆E and a complex linear surjective isometry θ:E→E\theta:E\to Eθ:E→E such that θ2=I\theta^2=Iθ2=I and Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for every v∈H+v\in H_+v∈H+​. A reflection group for rrr consists of complex linear surjective isometries Ut:E→EU_t:E\to EUt​:E→E, indexed by all t∈Rt\in\mathbb Rt∈R, such that U0x=xU_0x=xU0​x=x and Us+tx=Us(Utx)U_{s+t}x=U_s(U_tx)Us+t​x=Us​(Ut​x) for every s,t∈Rs,t\in\mathbb Rs,t∈R and x∈Ex\in Ex∈E; for each x∈Ex\in Ex∈E, the map t↦Utxt\mapsto U_txt↦Ut​x is continuous; Utx∈H+U_tx\in H_+Ut​x∈H+​ whenever t≥0t\geq0t≥0 and x∈H+x\in H_+x∈H+​, including t=0t=0t=0; and θ(Ut(θx))=U−tx\theta(U_t(\theta x))=U_{-t}xθ(Ut​(θx))=U−t​x for every t∈Rt\in\mathbb Rt∈R and x∈Ex\in Ex∈E.

QAlgorithms.Papers.Independent.Intakethird100.P92.domainQ

Let EEE be a complex inner product space, and let rrr consist of a closed complex linear subspace H+H_+H+​ and a complex linear surjective isometry θ\thetaθ satisfying θ2=I\theta^2=Iθ2=I and Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for every v∈H+v\in H_+v∈H+​. Let Hr\mathcal H_rHr​ be the separated completion of H+H_+H+​ with form [v,w]=⟨θv,w⟩[v,w]=\langle\theta v,w\rangle[v,w]=⟨θv,w⟩ and seminorm Re⁡⟨θv,v⟩\sqrt{\operatorname{Re}\langle\theta v,v\rangle}Re⟨θv,v⟩​, with canonical complex linear map q:H+→Hrq:H_+\to\mathcal H_rq:H+​→Hr​. For every complex linear subspace DDD of H+H_+H+​ and every v∈Dv\in Dv∈D, the domain map returns q(v)q(v)q(v) regarded as an element of the image subspace q(D)⊆Hrq(D)\subseteq\mathcal H_rq(D)⊆Hr​; its membership in that subspace is witnessed by vvv itself.

QAlgorithms.Papers.Independent.Intakethird100.P92.logReciprocal

For every nonnegative real number yyy, logReciprocal⁡(y)\operatorname{logReciprocal}(y)logReciprocal(y) is the extended nonnegative real defined by

logReciprocal⁡(y)={+∞,y=0,max⁡{log⁡(1/y),0},y>0,\operatorname{logReciprocal}(y)= \begin{cases} +\infty,&y=0,\\ \max\{\log(1/y),0\},&y>0, \end{cases}logReciprocal(y)={+∞,max{log(1/y),0},​y=0,y>0,​

where the finite value in the second branch is embedded into the extended nonnegative reals. In particular, it equals log⁡(1/y)\log(1/y)log(1/y) for 0<y<10<y<10<y<1, and it equals zero for y≥1y\geq1y≥1.

QAlgorithms.Papers.Independent.Intakethird100.P92.orbitSpan

For the ambient complex inner product space EEE, let (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ be a family of complex-linear isometric equivalences E→EE\to EE→E satisfying U0x=xU_0x=xU0​x=x and Us+tx=Us(Utx)U_{s+t}x=U_s(U_tx)Us+t​x=Us​(Ut​x) for every s,t∈Rs,t\in\mathbb Rs,t∈R and x∈Ex\in Ex∈E. For every complex subspace F⊆EF\subseteq EF⊆E and every set S⊆RS\subseteq\mathbb RS⊆R, the declaration defines the complex subspace span⁡C{Utv:t∈S, v∈F}\operatorname{span}_{\mathbb C}\{U_tv:t\in S,\ v\in F\}spanC​{Ut​v:t∈S, v∈F}. This is the algebraic span, consisting of finite complex linear combinations of the indicated vectors. The set SSS may be empty, in which case the resulting subspace is {0}\{0\}{0}.

QAlgorithms.Papers.Independent.Intakethird100.P92.osCompleteSpace

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E such that θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. Let HrH_rHr​ be the uniform completion of a copy of PPP carrying the seminorm x↦Re⁡⟨θx,x⟩Ex\mapsto\sqrt{\operatorname{Re}\langle\theta x,x\rangle_E}x↦Re⟨θx,x⟩E​​, with points at seminorm distance zero identified. This instance supplies completeness of HrH_rHr​: every Cauchy filter on HrH_rHr​ converges.

QAlgorithms.Papers.Independent.Intakethird100.P92.osCore

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for all v∈Pv\in Pv∈P, this defines a pre-inner-product core on records storing vectors of PPP by B(x,y)=⟨θvx,vy⟩EB(x,y)=\langle\theta v_x,v_y\rangle_EB(x,y)=⟨θvx​,vy​⟩E​, where vx,vyv_x,v_yvx​,vy​ are the stored vectors. It supplies the identities B(y,x)‾=B(x,y)\overline{B(y,x)}=B(x,y)B(y,x)​=B(x,y), B(x+y,z)=B(x,z)+B(y,z)B(x+y,z)=B(x,z)+B(y,z)B(x+y,z)=B(x,z)+B(y,z), and B(cx,y)=c‾ B(x,y)B(cx,y)=\overline c\,B(x,y)B(cx,y)=cB(x,y) for all records x,y,zx,y,zx,y,z and all c∈Cc\in\mathbb Cc∈C, together with Re⁡B(x,x)≥0\operatorname{Re}B(x,x)\geq0ReB(x,x)≥0 for every xxx. No nondegeneracy requirement is included.

QAlgorithms.Papers.Independent.Intakethird100.P92.osExtension

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E satisfying θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. Let HrH_rHr​ be the uniform completion of a copy of PPP with seminorm x↦Re⁡⟨θx,x⟩Ex\mapsto\sqrt{\operatorname{Re}\langle\theta x,x\rangle_E}x↦Re⟨θx,x⟩E​​, and let qr:P→Hrq_r:P\to H_rqr​:P→Hr​ be the canonical complex-linear map. For every complex-linear map T:P→PT:P\to PT:P→P, the declaration defines a continuous complex-linear map Hr→HrH_r\to H_rHr​→Hr​ as follows: if there exists a continuous complex-linear map A:Hr→HrA:H_r\to H_rA:Hr​→Hr​ satisfying A(qr(x))=qr(Tx)A(q_r(x))=q_r(Tx)A(qr​(x))=qr​(Tx) for every x∈Px\in Px∈P, it selects such an AAA by classical choice; if no such AAA exists, it returns the zero operator. No continuity assumption on TTT or existence assumption on AAA is imposed.

QAlgorithms.Papers.Independent.Intakethird100.P92.osInnerProductSpace

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E satisfying θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. Give a copy of PPP the form [x,y]r=⟨θx,y⟩E[x,y]_r=\langle\theta x,y\rangle_E[x,y]r​=⟨θx,y⟩E​ and the seminorm Re⁡[x,x]r\sqrt{\operatorname{Re}[x,x]_r}Re[x,x]r​​. This instance equips its uniform completion HrH_rHr​ with the inherited complex inner product space structure. In particular, if qr(x)q_r(x)qr​(x) denotes the canonical image of x∈Px\in Px∈P, then ⟨qr(x),qr(y)⟩Hr=⟨θx,y⟩E\langle q_r(x),q_r(y)\rangle_{H_r}=\langle\theta x,y\rangle_E⟨qr​(x),qr​(y)⟩Hr​​=⟨θx,y⟩E​. The completion identifies vectors whose difference has seminorm zero, so the canonical map need not be injective.

QAlgorithms.Papers.Independent.Intakethird100.P92.osNormedAddCommGroup

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, let OS⁡(r)\operatorname{OS}(r)OS(r) be the uniform completion of the records storing vectors of PPP with seminorm ∥x∥=Re⁡⟨θvx,vx⟩E\|x\|=\sqrt{\operatorname{Re}\langle\theta v_x,v_x\rangle_E}∥x∥=Re⟨θvx​,vx​⟩E​​. This declaration equips OS⁡(r)\operatorname{OS}(r)OS(r) with the normed additive commutative group structure inherited from that completion. Addition, negation, and the norm extend those of the original seminormed group, and in the completion a vector has norm zero exactly when it is zero.

QAlgorithms.Papers.Independent.Intakethird100.P92.osNormedSpace

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, let OS⁡(r)\operatorname{OS}(r)OS(r) be the uniform completion of the records storing vectors of PPP, whose scalar multiplication acts on stored vectors and whose seminorm is ∥x∥=Re⁡⟨θvx,vx⟩E\|x\|=\sqrt{\operatorname{Re}\langle\theta v_x,v_x\rangle_E}∥x∥=Re⟨θvx​,vx​⟩E​​. This declaration equips OS⁡(r)\operatorname{OS}(r)OS(r) with the complex normed-space structure inherited from that completion: complex scalar multiplication extends the original scalar multiplication and satisfies ∥cz∥=∣c∣∥z∥\|cz\|=|c|\|z\|∥cz∥=∣c∣∥z∥ for every c∈Cc\in\mathbb Cc∈C and every z∈OS⁡(r)z\in\operatorname{OS}(r)z∈OS(r).

QAlgorithms.Papers.Independent.Intakethird100.P92.osNull

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E satisfying θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. Let HrH_rHr​ be the uniform completion of a copy of PPP with seminorm x↦Re⁡⟨θx,x⟩Ex\mapsto\sqrt{\operatorname{Re}\langle\theta x,x\rangle_E}x↦Re⟨θx,x⟩E​​, and let qr:P→Hrq_r:P\to H_rqr​:P→Hr​ send each vector to its canonical image in that completion. The declaration defines the complex subspace of EEE obtained by including ker⁡qr\ker q_rkerqr​ into EEE, namely {x∈E:x∈P and qr(x)=0}\{x\in E:x\in P\text{ and }q_r(x)=0\}{x∈E:x∈P and qr​(x)=0}. This includes zero and may include nonzero vectors of EEE.

QAlgorithms.Papers.Independent.Intakethird100.P92.osPreAddCommGroup

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry with Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for all v∈Pv\in Pv∈P, this equips the records storing vectors of PPP with an additive commutative group structure by transporting the structure on PPP through the bijection that extracts the stored vector. Explicitly, zero stores 000, the sum of records storing v,wv,wv,w stores v+wv+wv+w, and the negative of a record storing vvv stores −v-v−v.

QAlgorithms.Papers.Independent.Intakethird100.P92.osPreEquiv

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), with PPP closed, θ\thetaθ a bijective complex linear isometry satisfying θ2=id\theta^2=\mathrm{id}θ2=id, and Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, this defines a bijection between the records OSPreSpace⁡(r)\operatorname{OSPreSpace}(r)OSPreSpace(r) that store a single vector of PPP and PPP itself. The forward map extracts the stored vector, and the inverse places a vector into such a record; both compositions are the identity.

QAlgorithms.Papers.Independent.Intakethird100.P92.osPreInnerProductSpace

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, this equips the records storing vectors of PPP with the complex inner-product structure obtained from the form ⟨x,y⟩r=⟨θvx,vy⟩E\langle x,y\rangle_r=\langle\theta v_x,v_y\rangle_E⟨x,y⟩r​=⟨θvx​,vy​⟩E​. This form is conjugate symmetric, conjugate linear in the first variable, linear in the second, and has nonnegative real part on the diagonal; its associated seminorm satisfies ∥x∥2=Re⁡⟨θvx,vx⟩E\|x\|^2=\operatorname{Re}\langle\theta v_x,v_x\rangle_E∥x∥2=Re⟨θvx​,vx​⟩E​. The underlying space is seminormed, and zero seminorm is not required here to imply x=0x=0x=0.

QAlgorithms.Papers.Independent.Intakethird100.P92.osPreModule

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, this equips the records storing vectors of PPP with a complex module structure transported from PPP. For every c∈Cc\in\mathbb Cc∈C, scalar multiplication by ccc sends the record storing vvv to the record storing cvcvcv, with addition and zero also transported from PPP.

QAlgorithms.Papers.Independent.Intakethird100.P92.osPreNormedSpace

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, this supplies the complex normed-space compatibility structure on records storing vectors of PPP, using scalar multiplication on their stored vectors and the seminorm ∥x∥=Re⁡⟨θvx,vx⟩E\|x\|=\sqrt{\operatorname{Re}\langle\theta v_x,v_x\rangle_E}∥x∥=Re⟨θvx​,vx​⟩E​​ induced by the form ⟨θvx,vy⟩E\langle\theta v_x,v_y\rangle_E⟨θvx​,vy​⟩E​. In particular, ∥cx∥=∣c∣∥x∥\|cx\|=|c|\|x\|∥cx∥=∣c∣∥x∥ for every c∈Cc\in\mathbb Cc∈C and every record xxx. The underlying additive structure is seminormed, so this declaration does not require zero seminorm to imply that the stored vector is zero.

QAlgorithms.Papers.Independent.Intakethird100.P92.osPreSeminormedAddCommGroup

For an ambient complex inner product space EEE and reflection data r=(P,θ)r=(P,\theta)r=(P,θ), where PPP is a closed complex linear subspace and θ\thetaθ is an involutive bijective complex linear isometry satisfying Re⁡⟨θv,v⟩E≥0\operatorname{Re}\langle\theta v,v\rangle_E\geq0Re⟨θv,v⟩E​≥0 for every v∈Pv\in Pv∈P, this equips the additive commutative group of records storing vectors of PPP with the seminormed structure induced by B(x,y)=⟨θvx,vy⟩EB(x,y)=\langle\theta v_x,v_y\rangle_EB(x,y)=⟨θvx​,vy​⟩E​. Its seminorm is ∥x∥=Re⁡⟨θvx,vx⟩E\|x\|=\sqrt{\operatorname{Re}\langle\theta v_x,v_x\rangle_E}∥x∥=Re⟨θvx​,vx​⟩E​​, and its distance is d(x,y)=Re⁡⟨θ(vx−vy),vx−vy⟩Ed(x,y)=\sqrt{\operatorname{Re}\langle\theta(v_x-v_y),v_x-v_y\rangle_E}d(x,y)=Re⟨θ(vx​−vy​),vx​−vy​⟩E​​. Distinct records may have distance zero.

QAlgorithms.Papers.Independent.Intakethird100.P92.osQ

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E satisfying θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. Form a copy PpreP_{\mathrm{pre}}Ppre​ of PPP, with its transported complex vector space operations and form [x,y]r=⟨θx,y⟩E[x,y]_r=\langle\theta x,y\rangle_E[x,y]r​=⟨θx,y⟩E​, and let HrH_rHr​ be its uniform completion for the resulting seminorm Re⁡[x,x]r\sqrt{\operatorname{Re}[x,x]_r}Re[x,x]r​​. The declaration defines the complex-linear map qr:P→Hrq_r:P\to H_rqr​:P→Hr​ by first placing xxx in its corresponding record in PpreP_{\mathrm{pre}}Ppre​ and then applying the canonical map into the completion. Vectors differing by a vector of seminorm zero have the same image.

QAlgorithms.Papers.Independent.Intakethird100.P92.polynomialValue

For every natural number ddd, every polynomial ppp with complex coefficients in variables indexed by the ddd-element set {0,…,d−1}\{0,\ldots,d-1\}{0,…,d−1}, and every vector xxx in real Euclidean ddd-space, polynomialValue⁡(p,x)\operatorname{polynomialValue}(p,x)polynomialValue(p,x) is the complex number obtained by evaluating each variable indexed by iii at the real coordinate xix_ixi​, embedded into C\mathbb CC:

polynomialValue⁡(p,x)=p(x0,…,xd−1).\operatorname{polynomialValue}(p,x) =p(x_0,\ldots,x_{d-1}).polynomialValue(p,x)=p(x0​,…,xd−1​).

The definition includes d=0d=0d=0, when there are no variables and evaluation returns the polynomial’s constant value at the unique vector in zero-dimensional Euclidean space.

QAlgorithms.Papers.Independent.Intakethird100.P92.positiveRestriction

Let EEE be a complex inner product space, and let rrr consist of a closed complex linear subspace H+H_+H+​ and a complex linear surjective isometry θ\thetaθ satisfying θ2=I\theta^2=Iθ2=I and Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for all v∈H+v\in H_+v∈H+​. Let UtU_tUt​, t∈Rt\in\mathbb Rt∈R, be complex linear surjective isometries satisfying U0=IU_0=IU0​=I, Us+t=UsUtU_{s+t}=U_sU_tUs+t​=Us​Ut​, continuity of every orbit t↦Utxt\mapsto U_txt↦Ut​x, preservation of H+H_+H+​ for all t≥0t\geq0t≥0, and θUtθ=U−t\theta U_t\theta=U_{-t}θUt​θ=U−t​. For each nonnegative real number ttt, including zero, the positive restriction is the complex linear map H+→H+H_+\to H_+H+​→H+​ sending vvv to UtvU_tvUt​v; its value belongs to H+H_+H+​ by the assumed preservation property.

QAlgorithms.Papers.Independent.Intakethird100.P92.projectionOn

For every complex linear subspace FFF of the ambient complex Hilbert space EEE, this defines the continuous complex linear map ΠF:E→E\Pi_F:E\to EΠF​:E→E obtained by first mapping EEE to FFF by the orthogonal projection when FFF is closed, or by the zero map when FFF is not closed, and then including FFF into EEE. Consequently, ΠF\Pi_FΠF​ is the ambient-valued orthogonal projection onto FFF in the closed case and the zero operator in the nonclosed case.

QAlgorithms.Papers.Independent.Intakethird100.P92.projectionTo

For a complex linear subspace FFF of the ambient complex Hilbert space EEE, this defines a continuous complex linear map PF:E→FP_F:E\to FPF​:E→F by a case distinction: if FFF is closed as a subset of EEE, use its resulting completeness to take the orthogonal projection onto FFF, with values regarded as elements of FFF; if FFF is not closed, take the zero map E→FE\to FE→F. Thus the definition applies to every subspace, including the zero subspace and nonclosed subspaces.

QAlgorithms.Papers.Independent.Intakethird100.P92.quantized

Let EEE be a complex inner product space, with a closed complex linear subspace H+H_+H+​ and a complex linear surjective isometry θ\thetaθ satisfying θ2=I\theta^2=Iθ2=I and Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for every v∈H+v\in H_+v∈H+​. Let Hr\mathcal H_rHr​ be the separated completion of H+H_+H+​ equipped with the sesquilinear form [v,w]=⟨θv,w⟩[v,w]=\langle\theta v,w\rangle[v,w]=⟨θv,w⟩ and seminorm Re⁡⟨θv,v⟩\sqrt{\operatorname{Re}\langle\theta v,v\rangle}Re⟨θv,v⟩​, and let q:H+→Hrq:H_+\to\mathcal H_rq:H+​→Hr​ be its canonical complex linear map; in particular, vectors at zero seminorm distance are identified. Let UtU_tUt​, t∈Rt\in\mathbb Rt∈R, be complex linear surjective isometries satisfying U0=IU_0=IU0​=I, Us+t=UsUtU_{s+t}=U_sU_tUs+t​=Us​Ut​, continuity of every orbit, UtH+⊆H+U_tH_+\subseteq H_+Ut​H+​⊆H+​ for t≥0t\geq0t≥0, and θUtθ=U−t\theta U_t\theta=U_{-t}θUt​θ=U−t​. For each nonnegative real ttt, the quantized operator Qt:Hr→HrQ_t:\mathcal H_r\to\mathcal H_rQt​:Hr​→Hr​ is defined as follows: if there exists a continuous complex linear map A:Hr→HrA:\mathcal H_r\to\mathcal H_rA:Hr​→Hr​ satisfying A(qv)=q(Utv)A(qv)=q(U_tv)A(qv)=q(Ut​v) for every v∈H+v\in H_+v∈H+​, then QtQ_tQt​ is a map chosen from those satisfying this condition; if no such map exists, QtQ_tQt​ is the zero operator.

QAlgorithms.Papers.Independent.Intakethird100.P92.quantizedFixed

Let EEE be a complex inner product space, with a closed complex linear subspace H+H_+H+​ and a complex linear surjective isometry θ\thetaθ satisfying θ2=I\theta^2=Iθ2=I and Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for all v∈H+v\in H_+v∈H+​. Write Hr\mathcal H_rHr​ for the separated completion of H+H_+H+​ with form [v,w]=⟨θv,w⟩[v,w]=\langle\theta v,w\rangle[v,w]=⟨θv,w⟩ and seminorm Re⁡⟨θv,v⟩\sqrt{\operatorname{Re}\langle\theta v,v\rangle}Re⟨θv,v⟩​, and q:H+→Hrq:H_+\to\mathcal H_rq:H+​→Hr​ for the canonical complex linear map. Let UtU_tUt​, t∈Rt\in\mathbb Rt∈R, be complex linear surjective isometries with U0=IU_0=IU0​=I, Us+t=UsUtU_{s+t}=U_sU_tUs+t​=Us​Ut​, continuous orbits, preservation of H+H_+H+​ for nonnegative times, and θUtθ=U−t\theta U_t\theta=U_{-t}θUt​θ=U−t​. For t≥0t\geq0t≥0, define QtQ_tQt​ to be a chosen continuous complex linear operator satisfying Qt(qv)=q(Utv)Q_t(qv)=q(U_tv)Qt​(qv)=q(Ut​v) for every v∈H+v\in H_+v∈H+​, if such an operator exists, and the zero operator otherwise. The quantized fixed subspace is the complex linear subspace

⋂t>0ker⁡(Qt−I)={x∈Hr: Qtx=x for every strictly positive real t}.\bigcap_{t>0}\ker(Q_t-I) =\{x\in\mathcal H_r:\ Q_tx=x\text{ for every strictly positive real }t\}.t>0⋂​ker(Qt​−I)={x∈Hr​: Qt​x=x for every strictly positive real t}.

The intersection is indexed by strictly positive times only.

QAlgorithms.Papers.Independent.Intakethird100.P92.quotientImage

Let EEE be a complex inner product space, and let rrr consist of a closed complex linear subspace H+⊆EH_+\subseteq EH+​⊆E and a complex linear surjective isometry θ:E→E\theta:E\to Eθ:E→E satisfying θ2=I\theta^2=Iθ2=I and Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for all v∈H+v\in H_+v∈H+​. Let Hr\mathcal H_rHr​ be the separated completion of H+H_+H+​ for the form [v,w]=⟨θv,w⟩[v,w]=\langle\theta v,w\rangle[v,w]=⟨θv,w⟩ and seminorm Re⁡⟨θv,v⟩\sqrt{\operatorname{Re}\langle\theta v,v\rangle}Re⟨θv,v⟩​, with canonical complex linear map q:H+→Hrq:H_+\to\mathcal H_rq:H+​→Hr​. For any complex linear subspace D⊆ED\subseteq ED⊆E, its quotient image is the set

{q(v):v∈H+ and v∈D}⊆Hr.\{q(v):v\in H_+\text{ and }v\in D\}\subseteq\mathcal H_r.{q(v):v∈H+​ and v∈D}⊆Hr​.

This is the image itself, without taking its closure, and DDD is not required to be contained in H+H_+H+​.

QAlgorithms.Papers.Independent.Intakethird100.P92.stieltjesDensity

For every natural number ddd, every measure ρ\rhoρ on [0,∞)[0,\infty)[0,∞), and every vector xxx in real Euclidean ddd-space, the function stieltjesDensity⁡(d,ρ,x)\operatorname{stieltjesDensity}(d,\rho,x)stieltjesDensity(d,ρ,x) is the extended nonnegative real

fd,ρ(x)=∫[0,∞)(y2+∥x∥2)−1 dρ(y),f_{d,\rho}(x)= \int_{[0,\infty)} \bigl(y^2+\|x\|^2\bigr)^{-1}\,d\rho(y),fd,ρ​(x)=∫[0,∞)​(y2+∥x∥2)−1dρ(y),

where yyy is embedded into the extended nonnegative reals and ∥x∥2\|x\|^2∥x∥2 is embedded using ofReal⁡(r)=max⁡(r,0)\operatorname{ofReal}(r)=\max(r,0)ofReal(r)=max(r,0). The integral may equal +∞+\infty+∞; the inverse is taken in the extended nonnegative reals, so the integrand is +∞+\infty+∞ when y=0y=0y=0 and x=0x=0x=0. The definition includes d=0d=0d=0, when the Euclidean space consists only of its zero vector, and it includes the zero measure.

QAlgorithms.Papers.Independent.Intakethird100.P92.stieltjesMeasure

For every natural number ddd and every measure ρ\rhoρ on [0,∞)[0,\infty)[0,∞), stieltjesMeasure⁡(d,ρ)\operatorname{stieltjesMeasure}(d,\rho)stieltjesMeasure(d,ρ) is the measure obtained by giving Lebesgue measure on real Euclidean ddd-space the extended nonnegative density

fd,ρ(x)=∫[0,∞)(y2+∥x∥2)−1 dρ(y).f_{d,\rho}(x)= \int_{[0,\infty)}\bigl(y^2+\|x\|^2\bigr)^{-1}\,d\rho(y).fd,ρ​(x)=∫[0,∞)​(y2+∥x∥2)−1dρ(y).

Thus, for every measurable set AAA,

stieltjesMeasure⁡(d,ρ)(A)=∫Afd,ρ(x) dx.\operatorname{stieltjesMeasure}(d,\rho)(A) =\int_A f_{d,\rho}(x)\,dx.stieltjesMeasure(d,ρ)(A)=∫A​fd,ρ​(x)dx.

The arithmetic is in the extended nonnegative reals, with 0−1=+∞0^{-1}=+\infty0−1=+∞; neither finiteness of the density nor finiteness of the resulting measure is assumed. The definition includes d=0d=0d=0 and the zero measure ρ\rhoρ.

QAlgorithms.Papers.Independent.Intakethird100.P92.timeZero

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E satisfying θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. The declaration defines the complex subspace P∩ker⁡(θ−id⁡E)P\cap\ker(\theta-\operatorname{id}_E)P∩ker(θ−idE​), consisting exactly of those x∈Ex\in Ex∈E such that x∈Px\in Px∈P and θx=x\theta x=xθx=x. This subspace contains zero and is allowed to equal {0}\{0\}{0}.

QAlgorithms.Papers.Independent.Intakethird100.P92.timeZeroCompression

Let EEE be a complex Hilbert space, let E+⊆EE_+\subseteq EE+​⊆E be a closed complex linear subspace, and let θ\thetaθ be an involutive complex linear isometric bijection with Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for every v∈E+v\in E_+v∈E+​. Let (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ be complex linear isometric bijections satisfying U0=idU_0=\mathrm{id}U0​=id, Us+t=UsUtU_{s+t}=U_sU_tUs+t​=Us​Ut​, continuity of every orbit map, preservation of E+E_+E+​ for t≥0t\geq0t≥0, and θUtθ=U−t\theta U_t\theta=U_{-t}θUt​θ=U−t​. Set Z={v∈E+:θv=v}Z=\{v\in E_+:\theta v=v\}Z={v∈E+​:θv=v}. For every real ttt, this defines a complex linear map Ct:Z→ZC_t:Z\to ZCt​:Z→Z by

Ct(v)=PZ(Utv),C_t(v)=P_Z(U_tv),Ct​(v)=PZ​(Ut​v),

where vvv is first included into EEE and PZ:E→ZP_Z:E\to ZPZ​:E→Z is the orthogonal projection onto the closed subspace ZZZ. The underlying projection convention is to use the zero map for a nonclosed target subspace; here ZZZ is closed. The time ttt may be negative, zero, or positive.

QAlgorithms.Papers.Independent.Intakethird100.P92.timeZeroQuotient

Let EEE be a complex inner product space, let E+⊆EE_+\subseteq EE+​⊆E be a closed complex linear subspace, and let θ:E→E\theta:E\to Eθ:E→E be an involutive complex linear isometric bijection satisfying Re⁡⟨θv,v⟩≥0\operatorname{Re}\langle\theta v,v\rangle\geq0Re⟨θv,v⟩≥0 for every v∈E+v\in E_+v∈E+​. Set Z={v∈E+:θv=v}Z=\{v\in E_+:\theta v=v\}Z={v∈E+​:θv=v}, and let HrH_rHr​ be the separated completion of a copy of E+E_+E+​ with the possibly degenerate inner product ⟨θv,w⟩\langle\theta v,w\rangle⟨θv,w⟩, with canonical complex linear map q:E+→Hrq:E_+\to H_rq:E+​→Hr​. This defines the complex linear map q0:Z→Hrq_0:Z\to H_rq0​:Z→Hr​ by including ZZZ into E+E_+E+​ and then applying qqq; explicitly, q0(v)=q(v)q_0(v)=q(v)q0​(v)=q(v) for every v∈Zv\in Zv∈Z. The definition itself makes no assertion of surjectivity.

QAlgorithms.Papers.Independent.Intakethird100.P92.toOSPre

For an ambient complex inner product space EEE, let rrr consist of a closed complex subspace P⊆EP\subseteq EP⊆E and a complex-linear isometric equivalence θ:E→E\theta:E\to Eθ:E→E satisfying θ2x=x\theta^2x=xθ2x=x for every x∈Ex\in Ex∈E and Re⁡⟨θx,x⟩E≥0\operatorname{Re}\langle\theta x,x\rangle_E\geq0Re⟨θx,x⟩E​≥0 for every x∈Px\in Px∈P. Let PpreP_{\mathrm{pre}}Ppre​ be the type whose elements are records containing one value in PPP, with addition and complex scalar multiplication transported from PPP through the map that extracts this value. The declaration defines the complex-linear map P→PpreP\to P_{\mathrm{pre}}P→Ppre​ sending each x∈Px\in Px∈P to the record containing xxx; it is the inverse of the value-extraction linear equivalence.

QAlgorithms.Papers.Independent.Intakethird100.P92.unitaryFixedSubspace

For the ambient complex inner product space EEE, let (Ut)t∈R(U_t)_{t\in\mathbb R}(Ut​)t∈R​ be a family of complex-linear isometric equivalences E→EE\to EE→E satisfying U0x=xU_0x=xU0​x=x and Us+tx=Us(Utx)U_{s+t}x=U_s(U_tx)Us+t​x=Us​(Ut​x) for every s,t∈Rs,t\in\mathbb Rs,t∈R and x∈Ex\in Ex∈E. The declaration defines the complex subspace ⋂t∈Rker⁡(Ut−id⁡E)\bigcap_{t\in\mathbb R}\ker(U_t-\operatorname{id}_E)⋂t∈R​ker(Ut​−idE​), consisting exactly of those x∈Ex\in Ex∈E satisfying Utx=xU_tx=xUt​x=x for every real ttt, including zero and negative times. It always contains zero and is allowed to be the zero subspace.

Human review
  • Endorsed by Shuze Chen · Nov 1, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Nov 1, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paper
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me