Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 3, (14)–(15), p. 7 — restricting R to the support of ρ and the bound B(ρ) ≤ A(ρ)

Proved
QReversal.NearOpt.eq_14_15

by mikedeng1 · Oct 15, 2026 · Mathlib 0df444a (Lean v4.33.1)

loewner-orderp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1petz-recoveryquantum-channel

Let E={pl,ρl}lE=\{p_l,\rho_l\}_lE={pl​,ρl​}l​ be an ensemble of commuting density matrices with average ρ=∑lplρl\rho=\sum_lp_l\rho_lρ=∑l​pl​ρl​, and let A∼{Ai}i\mathcal A\sim\{A_i\}_iA∼{Ai​}i​ be such that A(ρ)\mathcal A(\rho)A(ρ) has full support. Let R∼{Rk}k\mathcal R\sim\{R_k\}_kR∼{Rk​}k​ be trace-preserving and PCP_CPC​ the projector onto the support of ρ\rhoρ. Put

Rk′=PCRk,Bk=A(ρ)1/2Rk′†ρ−1/2,R'_k=P_CR_k,\qquad B_k=\mathcal A(\rho)^{1/2}R_k'^\dagger\rho^{-1/2},Rk′​=PC​Rk​,Bk​=A(ρ)1/2Rk′†​ρ−1/2,

with ρ−1/2\rho^{-1/2}ρ−1/2 the generalized inverse square root. Then:

  1. replacing RkR_kRk​ by Rk′R'_kRk′​ does not change the average entanglement fidelity: Fˉe(E,R′A)=Fˉe(E,RA)\bar F_e(E,\mathcal R'\mathcal A)=\bar F_e(E,\mathcal R\mathcal A)Fˉe​(E,R′A)=Fˉe​(E,RA);
  2. R′\mathcal R'R′ is trace-nonincreasing: ∑kRk′†Rk′≤I\sum_kR_k'^\dagger R'_k\le I∑k​Rk′†​Rk′​≤I;
  3. (14) Rk′=ρ1/2Bk†A(ρ)−1/2R'_k=\rho^{1/2}B_k^\dagger\mathcal A(\rho)^{-1/2}Rk′​=ρ1/2Bk†​A(ρ)−1/2 for every kkk;
  4. (15) B(ρ)≤A(ρ)\mathcal B(\rho)\le\mathcal A(\rho)B(ρ)≤A(ρ), where B∼{Bk}k\mathcal B\sim\{B_k\}_kB∼{Bk​}k​.

These are the opening reductions of the proof of Theorem 3: every recovery is rewritten in the same shape as the reversal operation, with BkB_kBk​ in place of AkA_kAk​, and the inequality (15) is what bounds the final sum in (34).

Formalization Note B\mathcal BB need not be trace-preserving. Part 1 needs no commutation; the commuting-ensemble hypothesis is the standing assumption of Theorem 3. Part 3 uses the full support of A(ρ)\mathcal A(\rho)A(ρ).

Preamble
import Mathlib
import Definitions.Def_PeresTerno_kraus_basics
import Definitions.Def_QReversal_NearOpt_Setting

open Matrix PeresTerno
open scoped MatrixOrder ComplexOrder
Formal statement
namespace QReversal.NearOpt

theorem eq_14_15 {n L ι κ : Type*} [Fintype n] [DecidableEq n] [Fintype L] [Fintype ι]
    [Fintype κ]
    (A : ι → Matrix n n ℂ) (p : L → ℝ) (ρs : L → Matrix n n ℂ)
    (hE : IsCommutingEnsemble p ρs) (ρ : Matrix n n ℂ) (hρ : ρ = avgState p ρs)
    (hfull : (krausUpdate A ρ).PosDef)
    (R : κ → Matrix n n ℂ) (hR : IsTracePreserving R) :
    let R' : κ → Matrix n n ℂ := fun k => supportProj ρ * R k
    let B : κ → Matrix n n ℂ := fun k => CFC.sqrt (krausUpdate A ρ) * (R' k)ᴴ * invSqrt ρ
    Fbar p ρs (comp R' A) = Fbar p ρs (comp R A) ∧
    IsTraceNonincreasing R' ∧
    (∀ k, R' k = CFC.sqrt ρ * (B k)ᴴ * invSqrt (krausUpdate A ρ)) ∧
    krausUpdate B ρ ≤ krausUpdate A ρ := by sorry

end QReversal.NearOpt
Source
Barnum, Knill, Reversing quantum dynamics with near-optimal quantum and classical fidelity, arXiv:quant-ph/0004088v2, p. 7, proof of Theorem 3, (14)–(15)
Human review
  • Endorsed by Shuze Chen · Oct 15, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 15, 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