Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Exact downconversion preserves weighted photon number and yields a diagonal signal

Open
QAlgorithms.Papers.Independent.Intakenext100.P33.downconversion_numberDiagonal

by Gohan · Oct 31, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Consider two full oscillator modes, pump aaa and signal bbb, with real coupling χ\chiχ and interaction Hamiltonian

HI=χ(a†b2+(b†)2a),W=2Na+Nb.H_I=\chi\bigl(a^\dagger b^2+(b^\dagger)^2a\bigr),\qquad W=2N_a+N_b.HI​=χ(a†b2+(b†)2a),W=2Na​+Nb​.

The Hamiltonian is constructed on the finite-occupation core with the usual square-root ladder coefficients and guarded lowering boundaries. It satisfies HIW=WHIH_IW=WH_IHI​W=WHI​ there. Every fixed-WWW sector has a finite Hermitian matrix, and for every real time ttt its matrix exponential e−itHIe^{-itH_I}e−itHI​ is unitary. Their explicitly constructed direct sum realizes a unitary on the full two-mode Hilbert space and preserves each weighted-number sector; square summability of its computed coefficients is a conclusion.

Let pm≥0p_m\ge0pm​≥0, ∑mpm=1\sum_mp_m=1∑m​pm​=1, with the series summable, be any pump photon-number distribution. The initial state is ∑mpm∣m⟩⟨m∣⊗∣0⟩⟨0∣\sum_mp_m|m\rangle\langle m|\otimes|0\rangle\langle0|∑m​pm​∣m⟩⟨m∣⊗∣0⟩⟨0∣. Write ψm(t)=U(t)∣m,0⟩\psi_m(t)=U(t)|m,0\rangleψm​(t)=U(t)∣m,0⟩. The computed reduced signal has the bounded occupation matrix

⟨n∣ρb(t)∣l⟩=∑m,j≥0pm ψm(t)j,nψm(t)j,l‾.\langle n|\rho_b(t)|l\rangle=\sum_{m,j\ge0}p_m\,\psi_m(t)_{j,n}\overline{\psi_m(t)_{j,l}}.⟨n∣ρb​(t)∣l⟩=m,j≥0∑​pm​ψm​(t)j,n​ψm​(t)j,l​​.

These sums converge, the resulting operator is positive trace class of trace one, and every off-diagonal entry vanishes. Its diagonal probability is ∑m,jpm∣ψm(t)j,n∣2\sum_{m,j}p_m|\psi_m(t)_{j,n}|^2∑m,j​pm​∣ψm​(t)j,n​∣2, with convergence asserted.

No finite-energy or finite-support assumption is imposed on the pump law, and no undepleted-pump approximation is used. Vacuum input or zero coupling gives vacuum signal.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P33
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P33

open Filter MeasureTheory

open scoped Topology Matrix ComplexConjugate

attribute [local instance] Classical.propDecidable

open QAlgorithms.Papers.Independent.Dependency50.P39

-- Accepted scalar phase; positive exponential, differing only by commuting scalars.

open QAlgorithms.Papers.Independent.Information100.P75 (phase)

/-- arXiv:quant-ph/0111157v1, PDF p.7, Eq.(32) and following paragraph.
The cubic Hamiltonian conserves 2Na+Nb on its core, and the constructed
finite-block direct sum is unitary and preserves weighted-number sectors.
Every normalized number-diagonal pump law, with signal vacuum, gives a
normalized trace-class number-diagonal signal. No finite moment or undepleted
pump approximation is imposed, and no commutation or output identity is assumed. -/
theorem downconversion_numberDiagonal (chi t : ℝ) :
    (∀ v : DownCore, downHamiltonian chi (weightedNumber v) =
      weightedNumber (downHamiltonian chi v)) ∧
    (∀ w, (downBlock chi w).IsHermitian ∧
      downBlockEvolution chi t w ∈ Matrix.unitaryGroup (Fin (w / 2 + 1)) ℂ) ∧
    (∀ v : BipartiteFock, Memℓp (downCoefficients chi t v) 2 ∧
      ∀ q, downEvolution chi t v q = downCoefficients chi t v q) ∧
    (∃ U : BipartiteFock ≃ₗᵢ[ℂ] BipartiteFock,
      ∀ v, U v = downEvolution chi t v) ∧
    (∀ (w : ℕ) (v : BipartiteFock),
      (∀ q, 2 * q.1 + q.2 ≠ w → v q = 0) →
      ∀ q, 2 * q.1 + q.2 ≠ w → downEvolution chi t v q = 0) ∧
    (∀ p : ℕ → ℝ, (∀ m, 0 ≤ p m) → Summable p → (∑' m, p m) = 1 →
      (∀ n l, Summable (fun k : ℕ × ℕ =>
        (p k.1 : ℂ) * evolvedPumpNumber chi t k.1 (k.2, n) *
          star (evolvedPumpNumber chi t k.1 (k.2, l)))) ∧
      HasFockMatrix (signalOutput p chi t) (signalMatrix p chi t) ∧
      IsDensity (signalOutput p chi t) ∧
      (∀ n l, n ≠ l → signalOutput p chi t (fockKet l) n = 0) ∧
      (∀ n, (signalOutput p chi t (fockKet n) n).re =
        ∑' k : ℕ × ℕ, p k.1 * Complex.normSq (evolvedPumpNumber chi t k.1 (k.2, n))) ∧
      (∀ n, Summable (fun k : ℕ × ℕ =>
        p k.1 * Complex.normSq (evolvedPumpNumber chi t k.1 (k.2, n))))) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P33

end
Source
The Quantum State of a Propagating Laser Field, arXiv:quant-ph/0111157v1, PDF p. 7, IIIC, Eq.(32) and following paragraph
Read-back

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

For every pair of real numbers χ,t\chi,tχ,t, let B=ℓ2(N2;C)\mathcal B=\ell^2(\mathbb N^2;\mathbb C)B=ℓ2(N2;C), let e(r,s)e_{(r,s)}e(r,s)​ be its standard basis, and consider also the vector space of finitely supported complex functions on N2\mathbb N^2N2. On this latter space, define We(r,s)=(2r+s)e(r,s)We_{(r,s)}=(2r+s)e_{(r,s)}We(r,s)​=(2r+s)e(r,s)​ and extend linearly the rule

Hχe(r,s)={χ(r+1)s(s−1) e(r+1,s−2),s≥2,0,s<2+{χr(s+1)(s+2) e(r−1,s+2),r≥1,0,r=0.H_\chi e_{(r,s)} = \begin{cases} \chi\sqrt{(r+1)s(s-1)}\,e_{(r+1,s-2)},&s\ge2,\\ 0,&s<2 \end{cases} + \begin{cases} \chi\sqrt{r(s+1)(s+2)}\,e_{(r-1,s+2)},&r\ge1,\\ 0,&r=0. \end{cases}Hχ​e(r,s)​={χ(r+1)s(s−1)​e(r+1,s−2)​,0,​s≥2,s<2​+{χr(s+1)(s+2)​e(r−1,s+2)​,0,​r≥1,r=0.​

For w∈Nw\in\mathbb Nw∈N, define a matrix HwH_wHw​, indexed by 0,…,⌊w/2⌋0,\ldots,\lfloor w/2\rfloor0,…,⌊w/2⌋, by

(Hw)ij={χ(j+1)(w−2j)(w−2j−1),i=j+1,χ(i+1)(w−2i)(w−2i−1),j=i+1,0,otherwise,Uw=exp⁡(−itHw),(H_w)_{ij}= \begin{cases} \chi\sqrt{(j+1)(w-2j)(w-2j-1)},&i=j+1,\\ \chi\sqrt{(i+1)(w-2i)(w-2i-1)},&j=i+1,\\ 0,&\text{otherwise}, \end{cases} \qquad U_w=\exp(-itH_w),(Hw​)ij​=⎩⎨⎧​χ(j+1)(w−2j)(w−2j−1)​,χ(i+1)(w−2i)(w−2i−1)​,0,​i=j+1,j=i+1,otherwise,​Uw​=exp(−itHw​),

with all natural-number subtractions truncated at zero. For v∈Bv\in\mathcal Bv∈B, define

cv(q1,q2)=∑j=0⌊w/2⌋(Uw)q1j v(j,w−2j),w=2q1+q2,c_v(q_1,q_2)= \sum_{j=0}^{\lfloor w/2\rfloor} (U_w)_{q_1j}\,v(j,w-2j), \qquad w=2q_1+q_2,cv​(q1​,q2​)=j=0∑⌊w/2⌋​(Uw​)q1​j​v(j,w−2j),w=2q1​+q2​,

and let Dχ,tvD_{\chi,t}vDχ,t​v be the vector with coefficients cvc_vcv​ when they are square summable, and zero otherwise. Then Hχ(Wv)=W(Hχv)H_\chi(Wv)=W(H_\chi v)Hχ​(Wv)=W(Hχ​v) for every finitely supported vvv; every HwH_wHw​ is Hermitian and every UwU_wUw​ is unitary; for every v∈Bv\in\mathcal Bv∈B, cvc_vcv​ is square summable and Dχ,tvD_{\chi,t}vDχ,t​v has exactly those coefficients; and there exists a surjective complex linear isometry U:B→BU:\mathcal B\to\mathcal BU:B→B satisfying Uv=Dχ,tvUv=D_{\chi,t}vUv=Dχ,t​v for every vvv. For every w∈Nw\in\mathbb Nw∈N and every v∈Bv\in\mathcal Bv∈B, if v(q1,q2)=0v(q_1,q_2)=0v(q1​,q2​)=0 whenever 2q1+q2≠w2q_1+q_2\ne w2q1​+q2​=w, then Dχ,tv(q1,q2)=0D_{\chi,t}v(q_1,q_2)=0Dχ,t​v(q1​,q2​)=0 whenever 2q1+q2≠w2q_1+q_2\ne w2q1​+q2​=w. Finally, for every p:N→Rp:\mathbb N\to\mathbb Rp:N→R satisfying p(m)≥0p(m)\ge0p(m)≥0 for every mmm, summability of ppp, and ∑mp(m)=1\sum_m p(m)=1∑m​p(m)=1, put vm=Dχ,te(m,0)v_m=D_{\chi,t}e_{(m,0)}vm​=Dχ,t​e(m,0)​, and define SSS to be a selected bounded operator on F=ℓ2(N;C)\mathcal F=\ell^2(\mathbb N;\mathbb C)F=ℓ2(N;C) with matrix

(Sel)n=∑(m,k)∈N2p(m) vm(k,n) vm(k,l)‾(Se_l)_n=\sum_{(m,k)\in\mathbb N^2} p(m)\,v_m(k,n)\,\overline{v_m(k,l)}(Sel​)n​=(m,k)∈N2∑​p(m)vm​(k,n)vm​(k,l)​

if one exists, and zero otherwise. For every n,l∈Nn,l\in\mathbb Nn,l∈N, this complex family is summable and SSS has exactly the displayed matrix. The operator SSS is a density operator in the following explicit sense: there exist sequences ur,wr∈Fu_r,w_r\in\mathcal Fur​,wr​∈F with ∑r∥ur∥∥wr∥<∞\sum_r\|u_r\|\|w_r\|<\infty∑r​∥ur​∥∥wr​∥<∞ and S=∑r∣ur⟩⟨wr∣S=\sum_r|u_r\rangle\langle w_r|S=∑r​∣ur​⟩⟨wr​∣ in operator norm, where ∣u⟩⟨w∣x=⟨w,x⟩u|u\rangle\langle w|x=\langle w,x\rangle u∣u⟩⟨w∣x=⟨w,x⟩u; also S∗=SS^*=SS∗=S, Re⁡⟨v,Sv⟩≥0\operatorname{Re}\langle v,Sv\rangle\ge0Re⟨v,Sv⟩≥0 for every vvv, its complex diagonal is summable, and its diagonal sum is one. Its off-diagonal entries vanish, and for every n∈Nn\in\mathbb Nn∈N,

Re⁡(Sen)n=∑(m,k)∈N2p(m)∣vm(k,n)∣2,\operatorname{Re}(Se_n)_n = \sum_{(m,k)\in\mathbb N^2}p(m)|v_m(k,n)|^2,Re(Sen​)n​=(m,k)∈N2∑​p(m)∣vm​(k,n)∣2,

with the real family on the right summable. No restriction on the signs or values of χ,t\chi,tχ,t is imposed; zero occupation numbers and w=0w=0w=0 are included, with H0H_0H0​ the one-dimensional zero matrix.

Human review
  • Endorsed by Shuze Chen · Oct 31, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 31, 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 paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me