Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Unnumbered conditional-entropy rederivation

Open
QAlgorithms.Papers.Independent.Support50.P14.gaussianConditionalEntropy

by Goku · Oct 23, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let AAA and BBB be nonempty finite collections of full oscillator modes, and let ρAB\rho_{AB}ρAB​ be a normalized density with finite complete moments. Let ρABG\rho^G_{AB}ρABG​ be a Gaussian density with the same symmetric covariance, allowing different first moments. All four entropies of the joint states and their AAA marginals are finite. With SSS measured in nats,

S(ρAB)−S(ρA)≤S(ρABG)−S(ρAG).S(\rho_{AB})-S(\rho_A)\le S(\rho^G_{AB})-S(\rho^G_A).S(ρAB​)−S(ρA​)≤S(ρABG​)−S(ρAG​).

Thus Gaussian states maximize conditional entropy at fixed covariance. The sign is opposite to the functional f(ρ)=S(ρA)−S(ρAB)f(\rho)=S(\rho_A)-S(\rho_{AB})f(ρ)=S(ρA​)−S(ρAB​) used by Lemma 1.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P14
Formal statement
noncomputable section

open Filter

open scoped BigOperators ENNReal Topology

attribute [local instance] Classical.propDecidable Classical.decEq

namespace QAlgorithms.Papers.Independent.Support50.P14

open QAlgorithms.RealRegisters (Hilbert)

open QAlgorithms.Papers.Independent.Dependency50.P33
  (State IsDensity IsTraceClass traceSize)

open QAlgorithms.Papers.Independent.Dependency50.P39 (displacementMatrix)

open QAlgorithms.Papers.Independent.Frontier30.P14
  (unitaryCoeff conjugateKernel boundedUnitaryConjugation)

open QAlgorithms.Papers.Independent.Frontier18.P07 (extendedEntropyNats)

open QAlgorithms.Papers.Independent.Claimed50.P47 (Occupation)

open QAlgorithms.Papers.Independent.Gap48.P20 (Copies)

/-- Conditional entropy rederived on PDF p.2: S(AB)-S(A) is MAXIMIZED by the
Gaussian counterpart. All four entropies are finite; toReal never hides infinity.
The input and comparator means may differ by local displacements. -/
theorem gaussianConditionalEntropy {A B : Type} [Fintype A] [Fintype B]
    [Nonempty A] [Nonempty B] (rho rhoG : State (Occupation (A ⊕ B)))
    (d dG : InterleavedQuadrature (A ⊕ B) → ℝ) (gamma : OscillatorCovariance (A ⊕ B))
    (hmoments : CompleteMoments rho.coefficients d gamma)
    (hgaussian : GaussianAt rhoG.coefficients dG gamma) :
    extendedEntropyNats rho.coefficients < ⊤ ∧
    extendedEntropyNats rhoG.coefficients < ⊤ ∧
    extendedEntropyNats (marginalFirst rho.coefficients) < ⊤ ∧
    extendedEntropyNats (marginalFirst rhoG.coefficients) < ⊤ ∧
    (extendedEntropyNats rho.coefficients).toReal -
      (extendedEntropyNats (marginalFirst rho.coefficients)).toReal ≤
    (extendedEntropyNats rhoG.coefficients).toReal -
      (extendedEntropyNats (marginalFirst rhoG.coefficients)).toReal := by sorry

end QAlgorithms.Papers.Independent.Support50.P14

end
Source
Extremality of Gaussian quantum states, arXiv:quant-ph/0509154v1, PDF p. 2, Unnumbered conditional-entropy rederivation
Read-back

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

Let A,BA,BA,B be finite, nonempty types, set I=A⊔BI=A\sqcup BI=A⊔B, NI=NI\mathcal N_I=\mathbb N^INI​=NI, and QI=I×{0,1}Q_I=I\times\{0,1\}QI​=I×{0,1}, and let ρ,ρG\rho,\rho_Gρ,ρG​ be states with coefficient kernels K,KGK,K_GK,KG​ on NI\mathcal N_INI​. A state kernel LLL here admits an expansion Luv=∑rar(u)br(v)‾L_{uv}=\sum_r a_r(u)\overline{b_r(v)}Luv​=∑r​ar​(u)br​(v)​ with ar,br∈ℓ2(NI;C)a_r,b_r\in\ell^2(\mathcal N_I;\mathbb C)ar​,br​∈ℓ2(NI​;C) and ∑r∥ar∥2∥br∥2<∞\sum_r\|a_r\|_2\|b_r\|_2<\infty∑r​∥ar​∥2​∥br​∥2​<∞, has positive semidefinite finite principal submatrices, and has summable real diagonal of sum 111. Let d,dG:QI→Rd,d_G:Q_I\to\mathbb Rd,dG​:QI​→R and let γ\gammaγ be a real matrix on QIQ_IQI​. Define

Dz(m,n)=e−∣z∣2/2m! n!∑j=0min⁡(m,n)zm−j(−z‾)n−jj!(m−j)!(n−j)!,Wξ(u,v)=∏i∈ID(−ξi,1+iξi,0)/2(ui,vi),D_z(m,n)=e^{-|z|^2/2}\sqrt{m!\,n!} \sum_{j=0}^{\min(m,n)} \frac{z^{m-j}(-\overline z)^{n-j}}{j!(m-j)!(n-j)!}, \quad W_\xi(u,v)=\prod_{i\in I} D_{(-\xi_{i,1}+i\xi_{i,0})/\sqrt2}(u_i,v_i),Dz​(m,n)=e−∣z∣2/2m!n!​j=0∑min(m,n)​j!(m−j)!(n−j)!zm−j(−z)n−j​,Wξ​(u,v)=i∈I∏​D(−ξi,1​+iξi,0​)/2​​(ui​,vi​), χL(ξ)=∑u∑vLuvWξ(v,u),me(ξ)=∑aξaea,Vγ(ξ)=∑a,bξaγabξb,E(L)=∑u[∑iui]+[Re⁡Luu]+,\chi_L(\xi)=\sum_u\sum_v L_{uv}W_\xi(v,u),\qquad m_e(\xi)=\sum_a\xi_a e_a,\qquad V_\gamma(\xi)=\sum_{a,b}\xi_a\gamma_{ab}\xi_b,\qquad \mathcal E(L)=\sum_u \Bigl[\sum_i u_i\Bigr]_+[\operatorname{Re}L_{uu}]_+,χL​(ξ)=u∑​v∑​Luv​Wξ​(v,u),me​(ξ)=a∑​ξa​ea​,Vγ​(ξ)=a,b∑​ξa​γab​ξb​,E(L)=u∑​[i∑​ui​]+​[ReLuu​]+​,

with [x]+=max⁡(x,0)[x]_+=\max(x,0)[x]+​=max(x,0) and the energy sum valued in [0,+∞][0,+\infty][0,+∞]. Assume E(K)<∞\mathcal E(K)<\inftyE(K)<∞, γT=γ\gamma^\mathsf T=\gammaγT=γ, and, for every ξ\xiξ, summability of each inner series in χK(ξ)\chi_K(\xi)χK​(ξ) and of the outer series of inner sums; also assume that t↦χK(tξ)t\mapsto\chi_K(t\xi)t↦χK​(tξ) has derivative i md(ξ)i\,m_d(\xi)imd​(ξ) at 000, and its derivative function has derivative −Vγ(ξ)/2−md(ξ)2-V_\gamma(\xi)/2-m_d(\xi)^2−Vγ​(ξ)/2−md​(ξ)2 there. Assume these same conditions for KGK_GKG​ with dGd_GdG​, together with

χKG(ξ)=exp⁡ ⁣(−Vγ(ξ)4+i mdG(ξ))for every ξ.\chi_{K_G}(\xi)= \exp\!\left(-\frac{V_\gamma(\xi)}4+i\,m_{d_G}(\xi)\right) \quad\text{for every }\xi.χKG​​(ξ)=exp(−4Vγ​(ξ)​+imdG​​(ξ))for every ξ.

No equality of ddd and dGd_GdG​ is required. Define the first marginal of any kernel LLL by

LA(u,v)=∑w∈NBLu⊔w, v⊔w,u,v∈NA.L_A(u,v)=\sum_{w\in\mathbb N^B} L_{u\sqcup w,\,v\sqcup w}, \qquad u,v\in\mathbb N^A.LA​(u,v)=w∈NB∑​Lu⊔w,v⊔w​,u,v∈NA.

For a kernel LLL on any label set, define

H(L)=sup⁡s finite[(Re⁡tr⁡Ls)log⁡(Re⁡tr⁡Ls)−Re⁡tr⁡(Lslog⁡(Ls))]+,H(L)=\sup_{s\text{ finite}} \left[ (\operatorname{Re}\operatorname{tr}L_s) \log(\operatorname{Re}\operatorname{tr}L_s) -\operatorname{Re}\operatorname{tr}(L_s\log(L_s)) \right]_+,H(L)=s finitesup​[(RetrLs​)log(RetrLs​)−Retr(Ls​log(Ls​))]+​,

where LsL_sLs​ is its finite principal compression, the logarithm is the real continuous functional calculus appearing in the code, scalar log⁡0=0\log0=0log0=0, and the empty compression contributes 000. The conclusion asserts finiteness of all four quantities H(K),H(KG),H(KA),H((KG)A)H(K),H(K_G),H(K_A),H((K_G)_A)H(K),H(KG​),H(KA​),H((KG​)A​), and the inequality

H(K)−H(KA)≤H(KG)−H((KG)A).H(K)-H(K_A)\le H(K_G)-H((K_G)_A).H(K)−H(KA​)≤H(KG​)−H((KG​)A​).

The subtraction is performed after converting these extended nonnegative entropies to real numbers; their asserted finiteness makes those conversions finite.

Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 24, 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