Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Thermal output and the three entropies

Open
QAlgorithms.Papers.Independent.Support50.P29.thermalOutputAndEntropies

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

prove2quantum

Let ℏ>0\hbar>0ℏ>0, N≥0N\ge0N≥0 be a finite mean photon number, and k≥0k\ge0k≥0 the amplitude coefficient. On the full one-mode Fock space let ρN\rho_NρN​ be the normalized thermal state with characteristic function exp⁡[−ℏ(N+1/2)∣z∣2/2]\exp[-\hbar(N+1/2)|z|^2/2]exp[−ℏ(N+1/2)∣z∣2/2], and let TkT_kTk​ be the quantum-limited attenuator or amplifier with a fixed vacuum environment. Define

N~={k2N,k≤1,k2N+k2−1,k>1.\widetilde N=\begin{cases}k^2N,&k\le1,\\k^2N+k^2-1,&k>1.\end{cases}N={k2N,k2N+k2−1,​k≤1,k>1.​

Then Tk(ρN)=ρN~T_k(\rho_N)=\rho_{\widetilde N}Tk​(ρN​)=ρN​. For g(x)=(x+1)log⁡(x+1)−xlog⁡xg(x)=(x+1)\log(x+1)-x\log xg(x)=(x+1)log(x+1)−xlogx, with 0log⁡0=00\log0=00log0=0, the actual von Neumann entropies in nats are finite and satisfy

H(ρN)=g(N),H(Tk(ρN))=g(N~),H(ρN,Tk)={g((1−k2)N),k≤1,g((k2−1)(N+1)),k>1.H(\rho_N)=g(N),\qquad H(T_k(\rho_N))=g(\widetilde N),\qquad H(\rho_N,T_k)=\begin{cases}g((1-k^2)N),&k\le1,\\g((k^2-1)(N+1)),&k>1.\end{cases}H(ρN​)=g(N),H(Tk​(ρN​))=g(N),H(ρN​,Tk​)={g((1−k2)N),g((k2−1)(N+1)),​k≤1,k>1.​

Entropy exchange is the spectral entropy of (Tk⊗Id)(ρ12)(T_k\otimes\mathrm{Id})(\rho_{12})(Tk​⊗Id)(ρ12​) for the normalized pure Schmidt purification, whose system marginal is ρN\rho_NρN​. The identity endpoint uses the convention stated on p.10. These formulas evaluate the three sides of the information triangle; they assert no capacity optimization.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P29
Formal statement
noncomputable section

open Filter

open scoped BigOperators ENNReal Topology ComplexOrder

attribute [local instance] Classical.propDecidable Classical.decEq

namespace QAlgorithms.Papers.Independent.Support50.P29

open QAlgorithms.RealRegisters (Hilbert)

open QAlgorithms.Papers.Independent.Dependency50.P10 (Phase sigma complexify)

open QAlgorithms.Papers.Independent.Dependency50.P33
  (State IsDensity rankOne productCoeff tensorApply IsTraceClassChannel)

open QAlgorithms.Papers.Independent.Frontier18.P18 (operatorCoeff)

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

open QAlgorithms.Papers.Independent.Frontier30.P10 (WeylSystem)

open QAlgorithms.Papers.Independent.Frontier30.P14 (partialRight PureDensity)

open QAlgorithms.Papers.Independent.Claimed50.P35
  (tracePair thermalInput thermalPurificationDensity vacuumChannel exchangeMatrix RealizedDensity)

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

open QAlgorithms.Papers.Independent.Information100.P49 (thermalEntropy)

/-- Holevo quant-ph/9809022v1, PDF pp.8-9, §4: the actual thermal output
and all three entropies in nats, in both channel regimes. The input exponent
uses hbar/2 as in (3.9), correcting the repeated hbar/4 misprint on p.9;
the amplifier is specified by (4.11), with the physical creation-mode noise. -/
theorem thermalOutputAndEntropies (N k hbar : ℝ)
    (hN : 0 ≤ N) (hk : 0 ≤ k) (hhbar : 0 < hbar) :
    RealizedDensity (thermalInput N) ∧
    RealizedDensity (vacuumChannel k (thermalInput N)) ∧
    PureDensity (thermalPurificationDensity N) ∧
    partialRight (thermalPurificationDensity N) = thermalInput N ∧
    RealizedDensity (exchangeOutput N k) ∧
    (∀ z : ℂ, QAlgorithms.Papers.Independent.Claimed50.P35.oneModeCharacteristic
      hbar (thermalInput N) z =
      Complex.exp (((-hbar * (N + 1 / 2) * Complex.normSq z / 2 : ℝ) : ℂ))) ∧
    vacuumChannel k (thermalInput N) = thermalInput (outputNumber N k) ∧
    extendedEntropyNats (thermalInput N) = ENNReal.ofReal (thermalEntropy N) ∧
    extendedEntropyNats (vacuumChannel k (thermalInput N)) =
      ENNReal.ofReal (thermalEntropy (outputNumber N k)) ∧
    extendedEntropyNats (exchangeOutput N k) = ENNReal.ofReal
      (thermalEntropy (if k ≤ 1 then (1 - k ^ 2) * N else (k ^ 2 - 1) * (N + 1))) := by sorry

end QAlgorithms.Papers.Independent.Support50.P29

end
Source
Sending Quantum Information with Gaussian States, arXiv:quant-ph/9809022v1, PDF p. 9, §4, unnumbered main result: thermal output and all three entropies
Read-back

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

For every real N,k,ℏN,k,\hbarN,k,ℏ satisfying N≥0N\geq0N≥0, k≥0k\geq0k≥0, and ℏ>0\hbar>0ℏ>0, make the following definitions. Let

tn(N)=22N+2(2N2N+2)n,(TN)mn=δmntn(N),t_n(N)=\frac{2}{2N+2}\left(\frac{2N}{2N+2}\right)^n,\qquad (T_N)_{mn}=\delta_{mn}t_n(N),tn​(N)=2N+22​(2N+22N​)n,(TN​)mn​=δmn​tn​(N),

and let HNH_NHN​ be the matrix on N2\mathbb N^2N2 with entry ta(N)tb(N)\sqrt{t_a(N)}\sqrt{t_b(N)}ta​(N)​tb​(N)​ at ((a,r),(b,t))((a,r),(b,t))((a,r),(b,t)) when a=ra=ra=r and b=tb=tb=t, and entry 000 otherwise. For ℓ,m,n∈N\ell,m,n\in\mathbb Nℓ,m,n∈N, let

Kℓ;mn={(m+ℓℓ)(1−k2)ℓkm,k≤1, n=m+ℓ,k−1(n+ℓℓ)(1−k−2)ℓ(k−1)n,k>1, m=n+ℓ,0,otherwise,K_{\ell;mn}= \begin{cases} \sqrt{\binom{m+\ell}{\ell}}(\sqrt{1-k^2})^\ell k^m, &k\leq1,\ n=m+\ell,\\ k^{-1}\sqrt{\binom{n+\ell}{\ell}} (\sqrt{1-k^{-2}})^\ell(k^{-1})^n, &k>1,\ m=n+\ell,\\ 0,&\text{otherwise}, \end{cases}Kℓ;mn​=⎩⎨⎧​(ℓm+ℓ​)​(1−k2​)ℓkm,k−1(ℓn+ℓ​)​(1−k−2​)ℓ(k−1)n,0,​k≤1, n=m+ℓ,k>1, m=n+ℓ,otherwise,​

and define

(VkX)ij=∑ℓ′∑a′Kℓ;ia(∑b′XabKℓ;jb‾),(EN,k)(i,r),(j,t)=∑ℓ′∑a′Kℓ;ia(∑b′(HN)(a,r),(b,t)Kℓ;jb‾).(\mathcal V_k X)_{ij} =\sum_\ell'\sum_a'K_{\ell;ia} \left(\sum_b'X_{ab}\overline{K_{\ell;jb}}\right), \qquad (E_{N,k})_{(i,r),(j,t)} =\sum_\ell'\sum_a'K_{\ell;ia} \left(\sum_b'(H_N)_{(a,r),(b,t)} \overline{K_{\ell;jb}}\right).(Vk​X)ij​=ℓ∑′​a∑′​Kℓ;ia​(b∑′​Xab​Kℓ;jb​​),(EN,k​)(i,r),(j,t)​=ℓ∑′​a∑′​Kℓ;ia​(b∑′​(HN​)(a,r),(b,t)​Kℓ;jb​​).

A matrix AAA on a label set LLL is a density here precisely when there are sequences ur,vr∈ℓ2(L;C)u_r,v_r\in\ell^2(L;\mathbb C)ur​,vr​∈ℓ2(L;C) with ∑r∥ur∥∥vr∥<∞\sum_r\|u_r\|\|v_r\|<\infty∑r​∥ur​∥∥vr​∥<∞ and Aij=∑r′ur(i)vr(j)‾A_{ij}=\sum_r'u_r(i)\overline{v_r(j)}Aij​=∑r′​ur​(i)vr​(j)​, every finite quadratic form ∑i,j∈Fzi‾Aijzj\sum_{i,j\in F}\overline{z_i}A_{ij}z_j∑i,j∈F​zi​​Aij​zj​ is real and nonnegative, and the real diagonal entries form a summable family whose sum is 111. Let B(A)B(A)B(A) be a chosen bounded complex linear operator on ℓ2(L;C)\ell^2(L;\mathbb C)ℓ2(L;C) having coordinate matrix AAA, if such an operator exists, and the zero operator otherwise; “realized density” additionally means (B(A)ej)i=Aij(B(A)e_j)_i=A_{ij}(B(A)ej​)i​=Aij​ for every i,ji,ji,j. For any matrix AAA, define

H(A)=sup⁡F⊆LF finitemax⁡ ⁣{0, (Re⁡tr⁡AF)log⁡(Re⁡tr⁡AF)−Re⁡tr⁡(AF cfc⁡(log⁡,AF))}∈[0,∞],\mathcal H(A)= \sup_{\substack{F\subseteq L\\F\ \mathrm{finite}}} \max\!\left\{0,\, (\operatorname{Re}\operatorname{tr}A_F) \log(\operatorname{Re}\operatorname{tr}A_F) -\operatorname{Re}\operatorname{tr} \bigl(A_F\,\operatorname{cfc}(\log,A_F)\bigr) \right\}\in[0,\infty],H(A)=F⊆LF finite​sup​max{0,(RetrAF​)log(RetrAF​)−Retr(AF​cfc(log,AF​))}∈[0,∞],

where AFA_FAF​ is the finite compression and cfc⁡\operatorname{cfc}cfc is the total continuous functional calculus for the real logarithm. Also put

g(x)=(x+1)log⁡(x+1)−xlog⁡x,o={k2N,k≤1,k2N+k2−1,k>1,e={(1−k2)N,k≤1,(k2−1)(N+1),k>1.g(x)=(x+1)\log(x+1)-x\log x,\qquad o= \begin{cases} k^2N,&k\leq1,\\ k^2N+k^2-1,&k>1, \end{cases} \qquad e= \begin{cases} (1-k^2)N,&k\leq1,\\ (k^2-1)(N+1),&k>1. \end{cases}g(x)=(x+1)log(x+1)−xlogx,o={k2N,k2N+k2−1,​k≤1,k>1,​e={(1−k2)N,(k2−1)(N+1),​k≤1,k>1.​

The theorem asserts all of the following together: TNT_NTN​ and Vk(TN)\mathcal V_k(T_N)Vk​(TN​) are realized densities; HNH_NHN​ is a density and there exists a unit vector u∈ℓ2(N2;C)u\in\ell^2(\mathbb N^2;\mathbb C)u∈ℓ2(N2;C) such that (HN)pq=upuq‾(H_N)_{pq}=u_p\overline{u_q}(HN​)pq​=up​uq​​; ∑b′(HN)(i,b),(j,b)=(TN)ij\sum_b'(H_N)_{(i,b),(j,b)}=(T_N)_{ij}∑b′​(HN​)(i,b),(j,b)​=(TN​)ij​ for every i,ji,ji,j; and EN,kE_{N,k}EN,k​ is a realized density. It further asserts

Vk(TN)=To,H(TN)=g(N),H(Vk(TN))=g(o),H(EN,k)=g(e),\mathcal V_k(T_N)=T_o,\qquad \mathcal H(T_N)=g(N),\qquad \mathcal H(\mathcal V_k(T_N))=g(o),\qquad \mathcal H(E_{N,k})=g(e),Vk​(TN​)=To​,H(TN​)=g(N),H(Vk​(TN​))=g(o),H(EN,k​)=g(e),

where the real entropy values on the right are embedded into [0,∞][0,\infty][0,∞] by taking their nonnegative parts. Finally, for every z∈Cz\in\mathbb Cz∈C, it asserts

∑m′∑n′(TN)mnDnm ⁣(iℏ/2 z)=exp⁡ ⁣(−ℏ(N+12)∣z∣22),\sum_m'\sum_n'(T_N)_{mn} D_{nm}\!\left(i\sqrt{\hbar/2}\,z\right) =\exp\!\left(-\frac{\hbar(N+\tfrac12)|z|^2}{2}\right),m∑′​n∑′​(TN​)mn​Dnm​(iℏ/2​z)=exp(−2ℏ(N+21​)∣z∣2​),

where

Dmn(ζ)=e−∣ζ∣2/2m! n!∑j=0min⁡(m,n)ζm−j(−ζ‾)n−jj! (m−j)! (n−j)!.D_{mn}(\zeta)=e^{-|\zeta|^2/2}\sqrt{m!\,n!} \sum_{j=0}^{\min(m,n)} \frac{\zeta^{m-j}(-\overline\zeta)^{n-j}} {j!\,(m-j)!\,(n-j)!}.Dmn​(ζ)=e−∣ζ∣2/2m!n!​j=0∑min(m,n)​j!(m−j)!(n−j)!ζm−j(−ζ​)n−j​.

The assertions include N=0N=0N=0, k=0k=0k=0, and k=1k=1k=1; powers with exponent 000 have value 111, including 000^000, and the total real logarithm has value 000 at 000.

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