Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

6.19

Open
QAlgorithms.Papers.Independent.Intakenext100.P07.smooth_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. Here FFF is squared algebraic fidelity, F∗F_*F∗​ is squared generalized fidelity, P=1−F∗P=\sqrt{1-F_*}P=1−F∗​​ is purified distance, and Δ(ρ,τ)=max⁡0≤M≤I∣Tr⁡((ρ−τ)M)∣\Delta(\rho,\tau)=\max_{0\le M\le I}|\operatorname{Tr}((\rho-\tau)M)|Δ(ρ,τ)=max0≤M≤I​∣Tr((ρ−τ)M)∣. The symbols Hmin⁡H_{\min}Hmin​ and Hmax⁡H_{\max}Hmax​ denote conditional min- and max-entropy, with values +∞+\infty+∞ and −∞-\infty−∞ at zero. Smoothing optimizes over all subnormalized states in the purified-distance ball; every entropy radius satisfies 0≤ε<Tr⁡ρ0\le\varepsilon<\sqrt{\operatorname{Tr}\rho}0≤ε<Trρ​.

For arbitrary CPTP F on B and sub-unital CPTP E on A, both Hmin^ε and Hmax^ε of the actual output (E⊗F)(ρ) are at least the input entropy.

Hε(A∣B)ρ≤Hε(A′∣B′)(E⊗F)(ρ)(H=Hmin⁡,Hmax⁡).H^\varepsilon(A|B)_\rho\le H^\varepsilon(A^\prime|B^\prime)_{(\mathcal E\otimes\mathcal F)(\rho)}\quad(H=H_{\min},H_{\max}).Hε(A∣B)ρ​≤Hε(A′∣B′)(E⊗F)(ρ)​(H=Hmin​,Hmax​).

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#6.19
6.19, arXiv:1504.00233v5, PDF p.104. -/
theorem smooth_data_processing [Nonempty A] [Nonempty B] [Nonempty C] [Nonempty D]
    (r : paperMat (A × B)) (hr : paperSubnormalized r) (e : ℝ) (he : Radius r e)
    (E : paperMat A →ₗ[ℂ] paperMat C) (F : paperMat B →ₗ[ℂ] paperMat D) (hE : paperCPTP E) (hF : paperCPTP F) (hu : paperBelow (E 1) 1) :
    paperSmoothMin r e ≤ paperSmoothMin (paperTensorAction E F r) e ∧ paperSmoothMax r e ≤ paperSmoothMax (paperTensorAction E F r) e := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P07

end
Source
Quantum Information Processing with Finite Resources -- Mathematical Foundations, arXiv:1504.00233v5, PDF p. 104, 6.19
Read-back

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

Let A,B,C,DA,B,C,DA,B,C,D be finite nonempty types with decidable equality, let r⪰0r\succeq0r⪰0 be a complex matrix on A×BA\times BA×B with τ(r)≤1\tau(r)\le1τ(r)≤1, where τ=Re⁡tr⁡\tau=\operatorname{Re}\operatorname{tr}τ=Retr, and let 0≤e<τ(r)0\le e<\sqrt{\tau(r)}0≤e<τ(r)​. Let E:Mat⁡(A,C)→Mat⁡(C,C)E:\operatorname{Mat}(A,\mathbb C)\to\operatorname{Mat}(C,\mathbb C)E:Mat(A,C)→Mat(C,C) and F:Mat⁡(B,C)→Mat⁡(D,C)F:\operatorname{Mat}(B,\mathbb C)\to\operatorname{Mat}(D,\mathbb C)F:Mat(B,C)→Mat(D,C) be complex-linear maps. Assume each preserves the complex trace of every matrix and is completely positive in the following explicit sense: for every natural number kkk, including k=0k=0k=0, applying the map to each matrix block of any positive semidefinite matrix on {0,…,k−1}×A\{0,\ldots,k-1\}\times A{0,…,k−1}×A, respectively {0,…,k−1}×B\{0,\ldots,k-1\}\times B{0,…,k−1}×B, produces a positive semidefinite matrix. Assume also IC−E(IA)⪰0I_C-E(I_A)\succeq0IC​−E(IA​)⪰0. Define ttt on C×DC\times DC×D by

t((c,d),(c′,d′))=∑a,a′,b,b′r((a,b),(a′,b′)) E(∣a⟩⟨a′∣)(c,c′) F(∣b⟩⟨b′∣)(d,d′).t((c,d),(c',d'))= \sum_{a,a',b,b'}r((a,b),(a',b'))\, E(|a\rangle\langle a'|)(c,c')\, F(|b\rangle\langle b'|)(d,d').t((c,d),(c′,d′))=a,a′,b,b′∑​r((a,b),(a′,b′))E(∣a⟩⟨a′∣)(c,c′)F(∣b⟩⟨b′∣)(d,d′).

For positive semidefinite matrices set R(u,v)=τ(u vu)R(u,v)=\tau(\sqrt{\sqrt u\,v\sqrt u})R(u,v)=τ(u​vu​​), with square roots from spectral functional calculus, and define Bη(v)={u⪰0:τ(u)≤1, 1−(R(u,v)+(1−τ(u))(1−τ(v)))2≤η}\mathcal B_\eta(v)=\{u\succeq0:\tau(u)\le1,\ \sqrt{1-(R(u,v)+\sqrt{(1-\tau(u))(1-\tau(v))})^2}\le\eta\}Bη​(v)={u⪰0:τ(u)≤1, 1−(R(u,v)+(1−τ(u))(1−τ(v))​)2​≤η}. On a bipartition U:VU:VU:V, let h−(q)=sup⁡{λ∈R:∃σ⪰0, τ(σ)≤1, 2−λIU⊗σ−q⪰0}h_-(q)=\sup\{\lambda\in\mathbb R:\exists\sigma\succeq0,\ \tau(\sigma)\le1,\ 2^{-\lambda}I_U\otimes\sigma-q\succeq0\}h−​(q)=sup{λ∈R:∃σ⪰0, τ(σ)≤1, 2−λIU​⊗σ−q⪰0}, h+(q)=sup⁡σ⪰0,τ(σ)≤1log⁡0(R(q,IU⊗σ)2)h_+(q)=\sup_{\sigma\succeq0,\tau(\sigma)\le1}\log_0(R(q,I_U\otimes\sigma)^2)h+​(q)=supσ⪰0,τ(σ)≤1​log0​(R(q,IU​⊗σ)2), H−η(q)=sup⁡u∈Bη(q)h−(u)H_-^\eta(q)=\sup_{u\in\mathcal B_\eta(q)}h_-(u)H−η​(q)=supu∈Bη​(q)​h−​(u), and H+η(q)=inf⁡u∈Bη(q)h+(u)H_+^\eta(q)=\inf_{u\in\mathcal B_\eta(q)}h_+(u)H+η​(q)=infu∈Bη​(q)​h+​(u); all extrema are extended-real, and log⁡0(0)=−∞\log_0(0)=-\inftylog0​(0)=−∞, while otherwise log⁡0=log⁡2\log_0=\log_2log0​=log2​. Then

H−e(r)≤H−e(t)andH+e(r)≤H+e(t),H_-^e(r)\le H_-^e(t) \qquad\text{and}\qquad H_+^e(r)\le H_+^e(t),H−e​(r)≤H−e​(t)andH+e​(r)≤H+e​(t),

using the bipartitions A:BA:BA:B and C:DC:DC:D, respectively; e=0e=0e=0 is allowed, while the radius hypothesis excludes zero trace.

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