Documentation

CsdLean4.LF3.Singlet.Kernel

LF3 Singlet / Kernel: pointer-sector kernel, correlation, marginals #

Category: 3-Local (LF3 algebraic core: P_st, cAmp, cst_squared_eq, correlation, marginals, no-signalling).

Paper §6 / §7.

The algebraic core: ‖c_{st}(a,b)‖² = (1 − st · a·b) / 4 (spec §9.8 calls this "the algebraic core of LF3"). Together with the strong-readout pointer probability P_{st}, the singlet correlation −a·b, the marginal identities = 1/2, and the operational no-signalling consequences, these supply the content of LF3_main_theorem.

cAmp is defined in closed form via Real.sqrt ((1 − st a·b)/4). This sidesteps the explicit construction of joint spin eigenstates |s_a, t_b⟩ (which require either a parametric spectral decomposition or a unit eigenvector of jointSpinProj); the squared-amplitude content ‖cAmp‖² = (1 − st a·b)/4 is preserved exactly. A future v2 may swap the closed-form definition for cAmp := inner ℂ jointSpinEig singlet once the eigenstate construction is added; downstream theorems consume only ‖cAmp‖², so the swap is transparent.

noncomputable def CSD.LF3.dotR (a b : DetectorSetting) :

Three-vector dot product a · b for two DetectorSettings, expanded as ∑_{i = 0, 1, 2} a_i b_i.

Equations
Instances For
    noncomputable def CSD.LF3.P_st (a b : DetectorSetting) (s t : Sign) :

    The pointer-sector probability P_{st}(a, b) = (1 − st · a·b) / 4 (paper §6.9, spec §9.8).

    Equations
    Instances For

      Non-negativity of P_st (needed for cAmp definition) #

      For unit vectors, |a · b| ≤ 1. Cauchy–Schwarz: identify dotR with the real inner product on EuclideanSpace ℝ (Fin 3) (the equality is rfl up to the ofLp and star_trivial reductions), then apply abs_real_inner_le_norm with the unit-norm hypotheses.

      theorem CSD.LF3.P_st_nonneg (a b : DetectorSetting) (s t : Sign) :
      0 P_st a b s t

      The pointer-sector probability is non-negative (needed to take Real.sqrt).

      Singlet amplitude (closed form) #

      noncomputable def CSD.LF3.cAmp (a b : DetectorSetting) (s t : Sign) :

      The singlet amplitude c_{st}(a, b), defined in closed form as the real square root of P_{st}(a, b). This is one canonical representative of the bra-ket form ⟨s_a, t_b | ψ⁻⟩; only ‖cAmp‖² is consumed downstream (in sectorVolume_strong_readout and the LF1↔LF2↔LF3 chain).

      Equations
      Instances For
        theorem CSD.LF3.cst_squared_eq (a b : DetectorSetting) (s t : Sign) :
        cAmp a b s t ^ 2 = P_st a b s t

        The algebraic core of LF3 (paper §6.9, spec §9.8): ‖c_{st}(a,b)‖² = (1 − st · a·b) / 4.

        Born-form equivalence (paper §6.11) #

        The closed-form cAmp = √P_st is a real-valued representative of the physical complex amplitude ⟨s_a, t_b | ψ⁻⟩. The Born-form fidelity claim is that this representative agrees with the genuine bra-ket inner product ⟨v, ψ⁻⟩ on its squared norm, whenever v : EuclideanSpace ℂ (Fin 2 × Fin 2) is an actual joint spin eigenstate |s_a, t_b⟩. The equivalence is expressed as a hypothesis on ‖inner ℂ v singlet‖² = P_st a b s t; in a v2 with a constructed jointSpinEig, this hypothesis discharges from the rank-1 projector identity jointSpinProj = |v⟩⟨v|.

        theorem CSD.LF3.cAmp_norm_sq_eq_inner_norm_sq (a b : DetectorSetting) (s t : Sign) (v : EuclideanSpace (Fin 2 × Fin 2)) (h_inner : inner v singlet ^ 2 = P_st a b s t) :

        Closed-form / bra-ket equivalence. If v : EuclideanSpace ℂ (Fin 2 × Fin 2) is a vector whose bra-ket inner product with the singlet has squared norm equal to P_{st}(a, b), then ‖cAmp s t (a, b)‖² = ‖⟨v, ψ⁻⟩‖². The hypothesis h_inner is the inner-product-norm equality that a rank-1 projector identity jointSpinProj = |v⟩⟨v| would entail (via a spectral argument); the theorem here consumes that conclusion directly. A v2 construction of jointSpinEig from the spectral decomposition of jointSpinProj would discharge h_inner automatically.

        Correlation and marginals (paper §6.10 / §7.3) #

        theorem CSD.LF3.correlation_eq_neg_dot (a b : DetectorSetting) :
        st : Sign × Sign, st.1.val * st.2.val * P_st a b st.1 st.2 = -dotR a b

        Singlet correlation at the strong-readout limit: ∑_{s,t} st·P_{st} = −a·b. Direct finite-Sign × Sign-sum calculation.

        theorem CSD.LF3.marginal_a_eq_half (a b : DetectorSetting) (s : Sign) :
        t : Sign, P_st a b s t = 1 / 2

        Singlet A-side marginal at the strong-readout limit: ∑_t P_{st} = 1/2. Two-term sum, 1/2 for each sign.

        theorem CSD.LF3.marginal_b_eq_half (a b : DetectorSetting) (t : Sign) :
        s : Sign, P_st a b s t = 1 / 2

        Singlet B-side marginal at the strong-readout limit.

        No-signalling corollaries (paper §7.10) #

        theorem CSD.LF3.no_signalling_strong_readout_a (a b b' : DetectorSetting) (s : Sign) :
        t : Sign, P_st a b s t = t : Sign, P_st a b' s t

        No-signalling, strong-readout, A side. Alice's marginal is independent of Bob's setting.

        theorem CSD.LF3.no_signalling_strong_readout_b (a a' b : DetectorSetting) (t : Sign) :
        s : Sign, P_st a b s t = s : Sign, P_st a' b s t

        No-signalling, strong-readout, B side. Bob's marginal is independent of Alice's setting.

        Bridge to the abstract layer (paper §6.11) #

        theorem CSD.LF3.abstract_sectorVolume_eq_P_st_at_singlet {K_A : Type u_1} {K_B : Type u_2} {H_SA : Type u_3} [NormedAddCommGroup K_A] [InnerProductSpace K_A] [FiniteDimensional K_A] [NormedAddCommGroup K_B] [InnerProductSpace K_B] [FiniteDimensional K_B] [NormedAddCommGroup H_SA] [InnerProductSpace H_SA] [FiniteDimensional H_SA] {S : SystemApparatusSetup K_A K_B H_SA} (P : ProjectorAlgebra S) (M : MeasurementUnitary S) (φA0 : K_A) (φB0 : K_B) (a b : DetectorSetting) (compat : StrongReadoutCompat P M φA0 φB0) (s t : Sign) :
        sectorVolume P (finalState M (cAmp a b) φA0 φB0) s t = P_st a b s t

        The strong-readout sector volume at the singlet equals the singlet kernel P_{st}(a, b). Composes sectorVolume_strong_readout (operator-level sector volume on a StrongReadoutCompat-equipped projector algebra and measurement unitary) with cst_squared_eq (the closed-form algebraic core for the singlet amplitude).