Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

5.5

Open
QAlgorithms.Papers.Independent.Intakenext100.P07.conditional_data_processing

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

prove2quantum

All quantum systems below are finite-dimensional complex Hilbert spaces of positive dimension, and logarithms are binary. A normalized state is positive semidefinite with trace one; a subnormalized state has trace at most one. Marginals are partial traces. Tildes denote sandwiched Rényi quantities and bars denote Petz quantities. Their trace functionals are Q~α=Tr⁡(σ(1−α)/(2α)ρσ(1−α)/(2α))α\widetilde Q_\alpha=\operatorname{Tr}(\sigma^{(1-\alpha)/(2\alpha)}\rho\sigma^{(1-\alpha)/(2\alpha)})^\alphaQ​α​=Tr(σ(1−α)/(2α)ρσ(1−α)/(2α))α and Q‾α=Tr⁡(ρασ1−α)\overline Q_\alpha=\operatorname{Tr}(\rho^\alpha\sigma^{1-\alpha})Q​α​=Tr(ρασ1−α); the divergences include division by Tr⁡ρ\operatorname{Tr}\rhoTrρ inside the logarithm. Matrix powers vanish on the kernel, including power zero. The notation ρ≪σ\rho\ll\sigmaρ≪σ means support inclusion. The upward conditional entropy optimizes the negative divergence over normalized reference states on the conditioning system; the downward entropy uses the actual marginal. The notation H(A∣B)=H(AB)−H(B)H(A|B)=H(AB)-H(B)H(A∣B)=H(AB)−H(B) denotes conditional von Neumann entropy.

For an actual CPTP map on B, each fixed and optimized Petz entropy (0≤α≤2) and sandwiched entropy (α≥1/2) cannot decrease; act as identity on A.

Hα(A∣B)ρ≤Hα(A∣B′)(idA⊗E)(ρ).H_\alpha(A|B)_\rho\le H_\alpha(A|B^\prime)_{(\mathrm{id}_A\otimes\mathcal E)(\rho)}.Hα​(A∣B)ρ​≤Hα​(A∣B′)(idA​⊗E)(ρ)​.

This result records the source’s operational or structural relation, with the stated endpoint and attainment clauses retained.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P07
Formal statement
noncomputable section

open Filter

open scoped BigOperators Matrix ComplexConjugate ComplexOrder Matrix.Norms.L2Operator Topology

namespace QAlgorithms.Papers.Independent.Intakenext100.P07

attribute [local instance] Classical.propDecidable Classical.decEq

variable {I J A B C D X Y Z : Type*}

variable [Fintype I] [DecidableEq I] [Fintype J] [DecidableEq J]

variable [Fintype A] [DecidableEq A] [Fintype B] [DecidableEq B]

variable [Fintype C] [DecidableEq C] [Fintype D] [DecidableEq D]

variable [Fintype X] [DecidableEq X] [Fintype Y] [DecidableEq Y]

variable [Fintype Z] [DecidableEq Z]

/-- NODE paper:1504.00233#5.5
5.5, arXiv:1504.00233v5, PDF p.81. -/
theorem conditional_data_processing [Nonempty A] [Nonempty B] [Nonempty C] (r : paperMat (A × B))
    (hr : IsDensityMatrix r) (E : paperMat B →ₗ[ℂ] paperMat C) (hE : paperCPTP E)
    (sand up : Bool) (a : ENNReal) (ha : DPIOrder sand a) :
    paperConditional sand up a r ≤ paperConditional sand up a (idTensorSuperop E r) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P07

end
Source
Quantum Information Processing with Finite Resources -- Mathematical Foundations, arXiv:1504.00233v5, PDF p. 81, 5.5
Read-back

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

Let A,B,CA,B,CA,B,C be finite, nonempty index types with decidable equality, let rrr be a positive semidefinite complex matrix on A×BA\times BA×B with trace 111, and let EEE be a complex-linear map from matrices on BBB to matrices on CCC that preserves complex trace on every matrix and whose every finite identity amplification preserves positive semidefiniteness. Choose independently one of the Petz or sandwiched formulas and one of the down or up formulas, and let α∈[0,+∞]\alpha\in[0,+\infty]α∈[0,+∞]; require α≤2\alpha\le2α≤2 for the Petz choice and α≥12\alpha\ge\tfrac12α≥21​ for the sandwiched choice. The quantities used in the assertion are defined as follows. For positive semidefinite X,YX,YX,Y, let M[p]M^{[p]}M[p] be functional calculus of xpx^pxp off zero with value zero at zero, let L(M)L(M)L(M) be functional calculus of log⁡2x\log_2 xlog2​x with its total value zero at zero, and set D(X∥Y)=Re⁡tr⁡(X(L(X)−L(Y)))/Re⁡tr⁡XD(X\|Y)=\operatorname{Re}\operatorname{tr}(X(L(X)-L(Y)))/\operatorname{Re}\operatorname{tr}XD(X∥Y)=Retr(X(L(X)−L(Y)))/RetrX if ker⁡Y⊆ker⁡X\ker Y\subseteq\ker XkerY⊆kerX, and +∞+\infty+∞ otherwise. For real ttt, put

QtP(X,Y)=Re⁡tr⁡(X[t]Y[1−t]),QtS(X,Y)=Re⁡tr⁡[(Y[(1−t)/(2t)]XY[(1−t)/(2t)])[t]].Q_t^{\mathrm P}(X,Y)=\operatorname{Re}\operatorname{tr}(X^{[t]}Y^{[1-t]}), \qquad Q_t^{\mathrm S}(X,Y)=\operatorname{Re}\operatorname{tr} \left[\left(Y^{[(1-t)/(2t)]}XY^{[(1-t)/(2t)]}\right)^{[t]}\right].QtP​(X,Y)=Retr(X[t]Y[1−t]),QtS​(X,Y)=Retr[(Y[(1−t)/(2t)]XY[(1−t)/(2t)])[t]].

For the chosen QtQ_tQt​, define R1=D(X∥Y)R_1=D(X\|Y)R1​=D(X∥Y); when t≠1t\ne1t=1, define Rt=(t−1)−1L(Qt(X,Y)/Re⁡tr⁡X)R_t=(t-1)^{-1}\mathcal L(Q_t(X,Y)/\operatorname{Re}\operatorname{tr}X)Rt​=(t−1)−1L(Qt​(X,Y)/RetrX) if (t<1 and Qt(X,Y)>0)(t<1\ \text{and}\ Q_t(X,Y)>0)(t<1 and Qt​(X,Y)>0) or ker⁡Y⊆ker⁡X\ker Y\subseteq\ker XkerY⊆kerX, and Rt=+∞R_t=+\inftyRt​=+∞ otherwise, where L(0)=−∞\mathcal L(0)=-\inftyL(0)=−∞ and L(x)=log⁡2x\mathcal L(x)=\log_2 xL(x)=log2​x otherwise. Define Dα=Rα\mathcal D_\alpha=R_\alphaDα​=Rα​ at finite positive α\alphaα, D0=inf⁡t>0Rt\mathcal D_0=\inf_{t>0}R_tD0​=inft>0​Rt​, and, at α=+∞\alpha=+\inftyα=+∞, define D+∞=sup⁡t>0Rt\mathcal D_{+\infty}=\sup_{t>0}R_tD+∞​=supt>0​Rt​ in the Petz case and D+∞=inf⁡{ℓ∈R:X≤2ℓY}\mathcal D_{+\infty}=\inf\{\ell\in\mathbb R:X\le2^\ell Y\}D+∞​=inf{ℓ∈R:X≤2ℓY} in the sandwiched case, with all infima and suprema taken in the extended reals and X≤YX\le YX≤Y meaning Y−XY-XY−X is positive semidefinite. For a matrix zzz on A×HA\times HA×H, put zH(h,h′)=∑az(a,h),(a,h′)z_H(h,h')=\sum_a z_{(a,h),(a,h')}zH​(h,h′)=∑a​z(a,h),(a,h′)​, and define

Hα↓(z)=−Dα(z∥1A⊗zH),Hα↑(z)=sup⁡τ≥0tr⁡τ=1[−Dα(z∥1A⊗τ)].H_\alpha^\downarrow(z)=-\mathcal D_\alpha(z\|1_A\otimes z_H), \qquad H_\alpha^\uparrow(z)= \sup_{\substack{\tau\ge0\\\operatorname{tr}\tau=1}} \bigl[-\mathcal D_\alpha(z\|1_A\otimes\tau)\bigr].Hα↓​(z)=−Dα​(z∥1A​⊗zH​),Hα↑​(z)=τ≥0trτ=1​sup​[−Dα​(z∥1A​⊗τ)].

Let r′r'r′ be the matrix on A×CA\times CA×C obtained by applying EEE to each BBB-indexed block of rrr, explicitly r(a,c),(a′,c′)′=E((r(a,b),(a′,b′))b,b′)c,c′r'_{(a,c),(a',c')}=E((r_{(a,b),(a',b')})_{b,b'})_{c,c'}r(a,c),(a′,c′)′​=E((r(a,b),(a′,b′)​)b,b′​)c,c′​. Then, with the same chosen formula and the same chosen arrow on both sides,

Hα↓ or ↑(r)≤Hα↓ or ↑(r′).H_\alpha^{\downarrow\text{ or }\uparrow}(r) \le H_\alpha^{\downarrow\text{ or }\uparrow}(r').Hα↓ or ↑​(r)≤Hα↓ or ↑​(r′).

This includes the order endpoints allowed by the respective inequalities, including α=0\alpha=0α=0 in the Petz case and α=+∞\alpha=+\inftyα=+∞ in the sandwiched case.

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