Original completed Green pair syntheses and covariance interfaces recovered from RG-0
DefinitionConnesGreen_RG0_original_actorsactual-zero-custodyconnes-weilrg0-dependency-integration
On the original completed Dirichlet carrier, retain the native bounded column synthesis, actual analytic zeta-zero multiplicities and reflected pair columns. The accepted RG-0 realization supplies exact physical column norms and square summability; the original positive, selected-negative, full-negative and complementary-negative synthesis operators are then defined by the unchanged native column-synthesis bodies. Basis-column identities and uniqueness give original actor custody. The positive covariance and complete complementary covariance retain every actual zero in their specified channels. No positivity, endpoint identification, arithmetic jump bound or RH is assumed.
Definition code
import Definitions.Def_ConnesGreen_actual_pair_columns
import Theorems.Thm_ConnesGreen_canonical_Green_realization_and_synthesis
set_option autoImplicit false
open Complex ConnesRZ ConnesRZFrontier
open scoped BigOperators InnerProductSpace lp ENNReal Classical
noncomputable section
namespace WeilDefect.ConnesNative
variable {ι H : Type*}
variable [NormedAddCommGroup H] [InnerProductSpace ℂ H] [CompleteSpace H]
/-- Auxiliary norm sequence for a supplied column family. This records a
summability proof; it does not change the columns or the physical carrier. -/
def columnNormCoefficients (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2)) : ℓ²(ι, ℂ) :=
⟨fun i => (‖v i‖ : ℂ), memℓp_gen (by
simpa [Complex.norm_real, Real.norm_eq_abs, Real.rpow_natCast] using hv)⟩
lemma columnSeries_norm_summable (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2))
(u : ℓ²(ι, ℂ)) : Summable (fun i => ‖u i • v i‖) := by
have hp : (2 : ℝ≥0∞).toReal.HolderConjugate (2 : ℝ≥0∞).toReal := by
simpa using Real.HolderConjugate.two_two
have hh := (lp.tsum_mul_le_mul_norm hp u (columnNormCoefficients v hv)).1
simpa [norm_smul, columnNormCoefficients, Complex.norm_real, Real.norm_eq_abs] using hh
lemma columnSeries_summable (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2))
(u : ℓ²(ι, ℂ)) : Summable (fun i => u i • v i) :=
(columnSeries_norm_summable v hv u).of_norm
lemma columnSeries_bound (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2))
(u : ℓ²(ι, ℂ)) :
‖∑' i, u i • v i‖ ≤ ‖columnNormCoefficients v hv‖ * ‖u‖ := by
have hp : (2 : ℝ≥0∞).toReal.HolderConjugate (2 : ℝ≥0∞).toReal := by
simpa using Real.HolderConjugate.two_two
have hh := lp.tsum_mul_le_mul_norm hp u (columnNormCoefficients v hv)
calc
_ ≤ ∑' i, ‖u i • v i‖ := norm_tsum_le_tsum_norm (columnSeries_norm_summable v hv u)
_ = ∑' i, ‖u i‖ * ‖columnNormCoefficients v hv i‖ := by
simp [norm_smul, columnNormCoefficients, Complex.norm_real, Real.norm_eq_abs]
_ ≤ ‖u‖ * ‖columnNormCoefficients v hv‖ := hh.2
_ = _ := mul_comm _ _
/-- Bounded synthesis of EXACTLY the supplied columns, with the ordinary l2
coefficient carrier. This is the extension of their finite linear combinations,
not a replacement for a native metric or a postulated bounded operator. -/
def columnSynthesis (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2)) :
ℓ²(ι, ℂ) →L[ℂ] H :=
LinearMap.mkContinuous
{ toFun := fun u => ∑' i, u i • v i
map_add' := fun u w => by
simp only [lp.coeFn_add, Pi.add_apply, add_smul]
exact (columnSeries_summable v hv u).tsum_add (columnSeries_summable v hv w)
map_smul' := fun c u => by
simp only [lp.coeFn_smul, Pi.smul_apply, smul_smul]
simpa only [smul_smul, smul_eq_mul, RingHom.id_apply] using
(columnSeries_summable v hv u).tsum_const_smul c }
‖columnNormCoefficients v hv‖ (columnSeries_bound v hv)
theorem columnSynthesis_apply (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2))
(u : ℓ²(ι, ℂ)) : columnSynthesis v hv u = ∑' i, u i • v i := rfl
theorem columnNormCoefficients_norm_sq (v : ι → H)
(hv : Summable (fun i => ‖v i‖ ^ 2)) :
‖columnNormCoefficients v hv‖ ^ 2 = ∑' i, ‖v i‖ ^ 2 := by
have hh := lp.norm_rpow_eq_tsum (by norm_num : 0 < (2 : ℝ≥0∞).toReal)
(columnNormCoefficients v hv)
simpa [columnNormCoefficients, Real.rpow_two, Complex.norm_real] using hh
theorem columnSynthesis_norm_sq_le (v : ι → H)
(hv : Summable (fun i => ‖v i‖ ^ 2)) :
‖columnSynthesis v hv‖ ^ 2 ≤ ∑' i, ‖v i‖ ^ 2 := by
have hb : ‖columnSynthesis v hv‖ ≤ ‖columnNormCoefficients v hv‖ := by
apply ContinuousLinearMap.opNorm_le_bound _ (norm_nonneg _)
exact columnSeries_bound v hv
rw [← columnNormCoefficients_norm_sq v hv]
exact pow_le_pow_left₀ (norm_nonneg _) hb 2
theorem columnSynthesis_single (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2))
(i : ι) (c : ℂ) : columnSynthesis v hv (lp.single 2 i c) = c • v i := by
classical
rw [columnSynthesis_apply]
rw [tsum_eq_single i]
· simp
· intro j hj
simp [lp.single_apply, Pi.single_apply, hj]
/-- The adjoint coordinates are the inner products with the ORIGINAL columns. -/
theorem columnSynthesis_adjoint_coordinate (v : ι → H)
(hv : Summable (fun i => ‖v i‖ ^ 2)) (h : H) (i : ι) :
(ContinuousLinearMap.adjoint (columnSynthesis v hv)) h i = ⟪v i, h⟫_ℂ := by
have hh := ContinuousLinearMap.adjoint_inner_right (columnSynthesis v hv)
(lp.single 2 i (1 : ℂ)) h
rw [columnSynthesis_single, one_smul] at hh
simpa only [lp.inner_single_left, RCLike.inner_apply, map_one, mul_one] using hh
/-- Parseval for the constructed adjoint: the exact column norm identity is
proved, not supplied as an arithmetic attachment premise. -/
theorem columnSynthesis_adjoint_norm_sq (v : ι → H)
(hv : Summable (fun i => ‖v i‖ ^ 2)) (h : H) :
‖(ContinuousLinearMap.adjoint (columnSynthesis v hv)) h‖ ^ 2 =
∑' i, ‖⟪v i, h⟫_ℂ‖ ^ 2 := by
have hh := lp.norm_rpow_eq_tsum (by norm_num : 0 < (2 : ℝ≥0∞).toReal)
((ContinuousLinearMap.adjoint (columnSynthesis v hv)) h)
simpa only [ENNReal.toReal_ofNat, Real.rpow_two,
columnSynthesis_adjoint_coordinate] using hh
/-- Custody/uniqueness: a supplied native synthesis with the same columns is
the constructed extension. No freedom to change the physical operator remains. -/
theorem columnSynthesis_unique [DecidableEq ι] (v : ι → H) (hv : Summable (fun i => ‖v i‖ ^ 2))
(S : ℓ²(ι, ℂ) →L[ℂ] H)
(hcolumns : ∀ i, S (lp.single 2 i (1 : ℂ)) = v i) : S = columnSynthesis v hv := by
ext u
have hs := S.hasSum (lp.hasSum_single (by norm_num : (2 : ℝ≥0∞) ≠ ⊤) u)
have heq : (fun i => S (lp.single 2 i (u i))) = fun i => u i • v i := by
funext i
have hh : lp.single (E := fun _ : ι => ℂ) 2 i (u i) =
u i • lp.single (E := fun _ : ι => ℂ) 2 i (1 : ℂ) := by
simpa using (lp.single_smul (E := fun _ : ι => ℂ) 2 i (u i) (1 : ℂ))
rw [hh, map_smul, hcolumns]
rw [heq] at hs
exact hs.tsum_eq.symm
end WeilDefect.ConnesNative
namespace WeilDefect.ConnesNative
open ConnesRZFrontier
variable {H : Type*} [NormedAddCommGroup H] [InnerProductSpace ℂ H]
theorem Green_pair_column_energy (v : CriticalZeros → H) (ρ : CriticalZeros) :
‖positiveGreenColumn v ρ‖ ^ 2 + ‖negativeGreenColumn v ρ‖ ^ 2 =
(‖weightedGreenColumn v ρ‖ ^ 2 + ‖weightedGreenColumn v (reflectedZero ρ)‖ ^ 2) / 2 := by
have hp := parallelogram_law_with_norm ℂ (weightedGreenColumn v ρ)
(weightedGreenColumn v (reflectedZero ρ))
unfold positiveGreenColumn negativeGreenColumn
simp only [norm_smul, mul_pow]
norm_num
nlinarith [hp]
end WeilDefect.ConnesNative
namespace ConnesGreen
open WeilDefect WeilDefect.ConnesNative
-- Same statement and pair normalization as native; the already accepted RG-0
-- aggregate supplies both the exact actual-zero metric and summability.
theorem canonical_actor_columns_summable (t : ℝ) (ht : 0 < t) :
Summable (fun ρ => ‖positiveGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ‖ ^ 2) ∧
Summable (fun ρ => ‖negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ‖ ^ 2) := by
have rg := canonical_Green_realization_and_synthesis t ht
let v := fun τ => sourceEmbed t (actualGreenSource τ)
have hs : Summable (fun ρ => ‖weightedGreenColumn v ρ‖ ^ 2) := by
have he : (fun ρ => ‖weightedGreenColumn v ρ‖ ^ 2) = weightedEnergy t := by
funext ρ
simp only [weightedGreenColumn, norm_smul, mul_pow, Complex.norm_real,
Real.norm_eq_abs, abs_of_nonneg (Real.sqrt_nonneg _),
Real.sq_sqrt (Nat.cast_nonneg _)]
change (zeroMult ρ.1 : ℝ) * ‖sourceEmbed t (actualGreenSource ρ)‖ ^ 2 = _
rw [(rg.2.1 ρ).2.2.2.2]
rfl
rw [he]
exact rg.2.2.2.1
have hr : Summable (fun ρ : CriticalZeros => ‖weightedGreenColumn v (reflectedZero ρ)‖ ^ 2) :=
hs.comp_injective (fun a b h => by
have hh := congrArg reflectedZero h
simpa only [reflectedZero_involutive] using hh)
have hb : Summable (fun ρ => ‖positiveGreenColumn v ρ‖ ^ 2 + ‖negativeGreenColumn v ρ‖ ^ 2) := by
simp_rw [Green_pair_column_energy]
exact (hs.add hr).div_const 2
exact ⟨Summable.of_nonneg_of_le (fun _ => sq_nonneg _)
(fun _ => le_add_of_nonneg_right (sq_nonneg _)) hb,
Summable.of_nonneg_of_le (fun _ => sq_nonneg _)
(fun _ => le_add_of_nonneg_left (sq_nonneg _)) hb⟩
def canonicalPositiveSynthesis (t : ℝ) (ht : 0 < t) :
ℓ²(CriticalZeros, ℂ) →L[ℂ] Physical t :=
columnSynthesis (positiveGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)))
(canonical_actor_columns_summable t ht).1
def canonicalSelectedSynthesis (t : ℝ) (ht : 0 < t) (S : Finset CriticalZeros) :
ℓ²({ρ : CriticalZeros // ρ ∈ S}, ℂ) →L[ℂ] Physical t :=
columnSynthesis (fun ρ => negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ.1)
((canonical_actor_columns_summable t ht).2.comp_injective Subtype.val_injective)
/-- Every original negative actor outside the selected packet, on the same carrier. -/
def canonicalBackgroundSynthesis (t : ℝ) (ht : 0 < t) (S : Finset CriticalZeros) :
ℓ²({ρ : CriticalZeros // ρ ∉ S}, ℂ) →L[ℂ] Physical t :=
columnSynthesis (fun ρ => negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ.1)
((canonical_actor_columns_summable t ht).2.comp_injective Subtype.val_injective)
def canonicalNegativeSynthesis (t : ℝ) (ht : 0 < t) :
ℓ²(CriticalZeros, ℂ) →L[ℂ] Physical t :=
columnSynthesis (negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)))
(canonical_actor_columns_summable t ht).2
theorem canonicalPositiveSynthesis_single (t : ℝ) (ht : 0 < t) (ρ : CriticalZeros) :
canonicalPositiveSynthesis t ht (lp.single 2 ρ (1 : ℂ)) =
positiveGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ := by
convert (columnSynthesis_single (positiveGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)))
(canonical_actor_columns_summable t ht).1 ρ (1 : ℂ)).trans (one_smul ℂ _) using 1
congr 1
ext τ
simp only [lp.single_apply, Pi.single_apply]
split_ifs <;> rfl
theorem canonicalSelectedSynthesis_single (t : ℝ) (ht : 0 < t) (S : Finset CriticalZeros)
(ρ : {ρ : CriticalZeros // ρ ∈ S}) :
canonicalSelectedSynthesis t ht S (lp.single 2 ρ (1 : ℂ)) =
negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ.1 := by
convert (columnSynthesis_single
(fun τ : {τ : CriticalZeros // τ ∈ S} =>
negativeGreenColumn (fun σ => sourceEmbed t (actualGreenSource σ)) τ.1)
((canonical_actor_columns_summable t ht).2.comp_injective Subtype.val_injective)
ρ (1 : ℂ)).trans (one_smul ℂ _) using 1
congr 1
ext τ
simp only [lp.single_apply, Pi.single_apply]
split_ifs <;> rfl
theorem canonicalPositiveSynthesis_unique (t : ℝ) (ht : 0 < t)
(T : ℓ²(CriticalZeros, ℂ) →L[ℂ] Physical t)
(hc : ∀ ρ, T (lp.single 2 ρ (1 : ℂ)) =
positiveGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ) :
T = canonicalPositiveSynthesis t ht := columnSynthesis_unique _ _ T hc
theorem canonicalSelectedSynthesis_unique (t : ℝ) (ht : 0 < t) (S : Finset CriticalZeros)
(T : ℓ²({ρ : CriticalZeros // ρ ∈ S}, ℂ) →L[ℂ] Physical t)
(hc : ∀ ρ, T (lp.single 2 ρ (1 : ℂ)) =
negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ.1) :
T = canonicalSelectedSynthesis t ht S := columnSynthesis_unique _ _ T hc
theorem canonicalBackgroundSynthesis_single (t : ℝ) (ht : 0 < t) (S : Finset CriticalZeros)
(ρ : {ρ : CriticalZeros // ρ ∉ S}) :
canonicalBackgroundSynthesis t ht S (lp.single 2 ρ (1 : ℂ)) =
negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ.1 := by
convert (columnSynthesis_single
(fun τ : {τ : CriticalZeros // τ ∉ S} =>
negativeGreenColumn (fun σ => sourceEmbed t (actualGreenSource σ)) τ.1)
((canonical_actor_columns_summable t ht).2.comp_injective Subtype.val_injective)
ρ (1 : ℂ)).trans (one_smul ℂ _) using 1
congr 1
ext τ
simp only [lp.single_apply, Pi.single_apply]
split_ifs <;> rfl
theorem canonicalBackgroundSynthesis_unique (t : ℝ) (ht : 0 < t) (S : Finset CriticalZeros)
(T : ℓ²({ρ : CriticalZeros // ρ ∉ S}, ℂ) →L[ℂ] Physical t)
(hc : ∀ ρ, T (lp.single 2 ρ (1 : ℂ)) =
negativeGreenColumn (fun τ => sourceEmbed t (actualGreenSource τ)) ρ.1) :
T = canonicalBackgroundSynthesis t ht S := columnSynthesis_unique _ _ T hc
def canonicalPositiveCovariance (t : ℝ) (ht : 0 < t) : Physical t →L[ℂ] Physical t :=
canonicalPositiveSynthesis t ht ∘L (canonicalPositiveSynthesis t ht).adjoint
def canonicalTailCovariance (t : ℝ) (ht : 0 < t) (F : Finset CriticalZeros) :
Physical t →L[ℂ] Physical t :=
canonicalBackgroundSynthesis t ht F ∘L (canonicalBackgroundSynthesis t ht F).adjoint
end ConnesGreen
Source
Recovered exact declaration bodies from monocap-tech/weil at 4ba3a3a569d72a0d5af2ba6ea320f948030dcca5: WeilDefect/Connes/ColumnSynthesis.lean, CanonicalGreenSeparation.lean, CanonicalGreenBackground.lean and CanonicalGreenMarker.lean. RG-0 is reused for the logically required metric and summability proofs; proof irrelevance leaves the synthesis values unchanged in the native environment.
Literature/source audit addendum (2026-10-11; COR-19): Exact native software custody, independently verified from retained local Git objects: monocap-tech/weil d73d0e7f1477c934a3a8f8b01012852128b977bb, WeilDefect/Connes/ColumnSynthesis.lean, SHA256 3528822f2a64896b3a325c80c783e7ef4be15cfff9a7a034db2753022a77c049; monocap-tech/weil d73d0e7f1477c934a3a8f8b01012852128b977bb, WeilDefect/Connes/CanonicalGreenSeparation.lean, SHA256 4e5b59019c8c7d26c87289597de2965d3eead091044f49a41af1783349c77c4b; monocap-tech/weil d73d0e7f1477c934a3a8f8b01012852128b977bb, WeilDefect/Connes/CanonicalGreenBackground.lean, SHA256 6c8a15517dfbdfee96d47da283627a8ec956bdca5a6f0f374df4959f93772f29; monocap-tech/weil d73d0e7f1477c934a3a8f8b01012852128b977bb, WeilDefect/Connes/CanonicalGreenMarker.lean, SHA256 cb38f6de94e8f5c299b73617bbf309df3cd605abee13ce1ce9ecb5a9c7c4408b. Public availability of these historical pins has not been checked; this addendum verifies local source bytes and preserves the original source ancestry. The unchanged graph RHS values were checked against native definitions by Lean rfl. This is source/value custody, not an assertion that contextual literature supplies the whole definition bundle.