Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 4

Open
QAlgorithms.Papers.Independent.Gap48.P18.strong_projection_capacity_and_average_approximation

by Goku · Oct 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let AAA be a nonempty state set and let orthogonal projections PnP_nPn​ converge strongly to the identity. No finite-rank, nesting, or monotonicity assumption is imposed. If Cˉ(A)<∞\bar C(A)<\inftyCˉ(A)<∞, normalization is eventually valid uniformly on AAA, image capacities are eventually finite, and

Cˉ(ΘPn(A))→Cˉ(A),Ω(ΘPn(A))→Ω(A) in trace norm.\bar C(\Theta_{P_n}(A))\to\bar C(A),\qquad\Omega(\Theta_{P_n}(A))\to\Omega(A)\ \text{in trace norm}.Cˉ(ΘPn​​(A))→Cˉ(A),Ω(ΘPn​​(A))→Ω(A) in trace norm.

Conversely, if some strongly exhaustive projection sequence has all normalization maps defined on AAA and uniformly bounded image capacities, then Cˉ(A)<∞\bar C(A)<\inftyCˉ(A)<∞. Arbitrary initial terms outside the positive-trace domain do not affect the first limits.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Gap48_P18
Formal statement
noncomputable section

open MeasureTheory Filter

open scoped ENNReal NNReal Topology Matrix.Norms.L2Operator

open QAlgorithms.Papers.Independent.Dependency50.P33
  (State PureState traceSize traceClosure IsTraceClassChannel productCoeff)

open QAlgorithms.Papers.Independent.Frontier18.P07
  (GeneralizedEnsemble HasFiniteSupport HasAverage averageCoefficients
   extendedEntropyNats extendedRelativeEntropyNats
   finiteEnsembleCapacity pureStates)

open QAlgorithms.Papers.Independent.Frontier18.P01 (traceFirst traceSecond)

namespace QAlgorithms.Papers.Independent.Gap48.P18

attribute [local instance] Classical.propDecidable Classical.decEq

/-- Theorem 4, arXiv:quant-ph/0510073v2, PDF pp.73.
Early empty images are totalized only for writing the average sequence. Eventual positive uniform mass, nonempty images, and finite image capacities are conclusions. -/
theorem strong_projection_capacity_and_average_approximation {I : Type}
    [Countable I] [Nonempty I] (A : Set (State I)) (hA : A.Nonempty)
    (P : ℕ → BoundedOrthogonalProjection I)
    (hstrong : ∀ v : QAlgorithms.RealRegisters.Hilbert I,
      Tendsto (fun n => (P n).operator v) atTop (𝓝 v)) :
    (setChiCapacity A < ⊤ →
      (∀ᶠ n in atTop, 0 < projectionSuccessInfimum (P n) A) ∧
      (∀ᶠ n in atTop, (projectionImage (P n) A).Nonempty ∧
        setChiCapacity (projectionImage (P n) A) < ⊤) ∧
      Tendsto (fun n => setChiCapacity (projectionImage (P n) A)) atTop (𝓝 (setChiCapacity A)) ∧
      Tendsto (fun n => projectedAverage (P n) A hA) atTop (𝓝 (optimalAverage A hA))) ∧
      ((∀ n, ∀ rho ∈ A, 0 < projectionMass (P n) rho) →
        (⨆ n, setChiCapacity (projectionImage (P n) A)) < ⊤ → setChiCapacity A < ⊤) := by sorry

end QAlgorithms.Papers.Independent.Gap48.P18

end
Source
Entropic characteristics of subsets of states, arXiv:quant-ph/0510073v2, PDF p. 73, Theorem 4
Read-back

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

Let III be any countable, nonempty type, and let HI=ℓ2(I;C)\mathcal H_I=\ell^2(I;\mathbb C)HI​=ℓ2(I;C). A state on III is a complex matrix X=(Xij)X=(X_{ij})X=(Xij​) admitting an expansion Xij=∑m=0∞um(i)vm(j)‾X_{ij}=\sum_{m=0}^{\infty}u_m(i)\overline{v_m(j)}Xij​=∑m=0∞​um​(i)vm​(j)​, where um,vm∈HIu_m,v_m\in\mathcal H_Ium​,vm​∈HI​ and ∑m∥um∥∥vm∥<∞\sum_m\|u_m\|\|v_m\|<\infty∑m​∥um​∥∥vm​∥<∞, such that every finite quadratic form ∑i,j∈szi‾Xijzj\sum_{i,j\in s}\overline{z_i}X_{ij}z_j∑i,j∈s​zi​​Xij​zj​, for every finite s⊆Is\subseteq Is⊆I and every z:I→Cz:I\to\mathbb Cz:I→C, is real and nonnegative, and such that ∑iRe⁡Xii\sum_i\operatorname{Re}X_{ii}∑i​ReXii​ is summable and equals 111. Write SI\mathcal S_ISI​ for these states, identifying each state with its coefficient matrix. For any matrix ZZZ, define

τ(Z)=inf⁡Zij=∑mum(i)vm(j)‾∑m∥um∥∥vm∥<∞[∑m∥um∥∥vm∥]+,[r]+=max⁡(r,0)∈[0,∞],\tau(Z)=\inf_{\substack{Z_{ij}=\sum_m u_m(i)\overline{v_m(j)}\\ \sum_m\|u_m\|\|v_m\|<\infty}} \left[\sum_m\|u_m\|\|v_m\|\right]_+, \qquad [r]_+=\max(r,0)\in[0,\infty],τ(Z)=Zij​=∑m​um​(i)vm​(j)​∑m​∥um​∥∥vm​∥<∞​inf​[m∑​∥um​∥∥vm​∥]+​,[r]+​=max(r,0)∈[0,∞],

where the infimum is +∞+\infty+∞ if there is no such expansion. Equip SI\mathcal S_ISI​ with the topology generated by all sets {Y:τ(Y−X)<ϵ}\{Y:\tau(Y-X)<\epsilon\}{Y:τ(Y−X)<ϵ}, with X∈SIX\in\mathcal S_IX∈SI​ and real ϵ>0\epsilon>0ϵ>0, and with its Borel measurable structure. To define the capacity used below, for finite complex matrices B,CB,CB,C on the same index set put

d(B,C)={[Re⁡tr⁡(Blog⁡B−Blog⁡C+C−B)]+,ker⁡C⊆ker⁡B,+∞,otherwise,D(X,Y)=sup⁡s⊆I finited(X∣s×s,Y∣s×s),d(B,C)= \begin{cases} \left[\operatorname{Re}\operatorname{tr} \bigl(B\log B-B\log C+C-B\bigr)\right]_+, &\ker C\subseteq\ker B,\\ +\infty,&\text{otherwise}, \end{cases} \qquad D(X,Y)=\sup_{s\subseteq I\ {\rm finite}}d(X|_{s\times s},Y|_{s\times s}),d(B,C)={[Retr(BlogB−BlogC+C−B)]+​,+∞,​kerC⊆kerB,otherwise,​D(X,Y)=s⊆I finitesup​d(X∣s×s​,Y∣s×s​),

where matrix logarithms use continuous functional calculus and the real logarithm has value 000 at 000; the supremum includes the empty compression. For a Borel probability measure μ\muμ on SI\mathcal S_ISI​, let (Xˉμ)ij=∫Xij dμ(X)(\bar X_\mu)_{ij}=\int X_{ij}\,d\mu(X)(Xˉμ​)ij​=∫Xij​dμ(X) and χ(μ)=∫−D(X,Xˉμ) dμ(X)\chi(\mu)=\int^-D(X,\bar X_\mu)\,d\mu(X)χ(μ)=∫−D(X,Xˉμ​)dμ(X), using the nonnegative extended integral; the coefficient integrals are the total complex integrals, which take value 000 when the corresponding function is not integrable. For any B⊆SIB\subseteq\mathcal S_IB⊆SI​, set

C(B)=sup⁡{χ(μ):μ is a Borel probability measure, supp⁡μ is finite, supp⁡μ⊆B}.C(B)=\sup\{\chi(\mu):\mu\text{ is a Borel probability measure}, \ \operatorname{supp}\mu\text{ is finite},\ \operatorname{supp}\mu\subseteq B\}.C(B)=sup{χ(μ):μ is a Borel probability measure, suppμ is finite, suppμ⊆B}.

This is a supremum in [0,∞][0,\infty][0,∞], with empty supremum 000, and the support is taken in the state topology just specified. Let A⊆SIA\subseteq\mathcal S_IA⊆SI​ be nonempty, with a given witness of nonemptiness. For each n∈Nn\in\mathbb Nn∈N, let Pn:HI→HIP_n:\mathcal H_I\to\mathcal H_IPn​:HI​→HI​ be a bounded complex linear operator satisfying Pn2=PnP_n^2=P_nPn2​=Pn​ and Pn∗=PnP_n^*=P_nPn∗​=Pn​, and suppose ∥Pnv−v∥→0\|P_nv-v\|\to0∥Pn​v−v∥→0 for every v∈HIv\in\mathcal H_Iv∈HI​. Writing eie_iei​ for the coordinate unit vectors, define

QP(X)ab=∑i∈I∑j∈I(Pei)aXij(Pej)b‾,pP(X)=∑a∈IRe⁡QP(X)aa,AP={Y∈SI:∃X∈A, pP(X)>0, Y=pP(X)−1QP(X)},qP(A)=inf⁡X∈A[pP(X)]+.Q_P(X)_{ab}=\sum_{i\in I}\sum_{j\in I}(Pe_i)_aX_{ij}\overline{(Pe_j)_b}, \quad p_P(X)=\sum_{a\in I}\operatorname{Re}Q_P(X)_{aa}, \quad A_P=\{Y\in\mathcal S_I:\exists X\in A,\ p_P(X)>0,\ Y=p_P(X)^{-1}Q_P(X)\}, \quad q_P(A)=\inf_{X\in A}[p_P(X)]_+.QP​(X)ab​=i∈I∑​j∈I∑​(Pei​)a​Xij​(Pej​)b​​,pP​(X)=a∈I∑​ReQP​(X)aa​,AP​={Y∈SI​:∃X∈A, pP​(X)>0, Y=pP​(X)−1QP​(X)},qP​(A)=X∈Ainf​[pP​(X)]+​.

These sums use the total infinite-sum convention, giving 000 to a nonsummable family. For a nonempty set BBB, let ωB\omega_BωB​ be a fixed classical choice of a state satisfying D(X,ωB)≤C(B)D(X,\omega_B)\le C(B)D(X,ωB​)≤C(B) for every X∈BX\in BX∈B, if any such state exists, and otherwise a state chosen from the given witness that BBB is nonempty. Let ω^P,A=ωAP\widehat\omega_{P,A}=\omega_{A_P}ωP,A​=ωAP​​ when APA_PAP​ is nonempty, and ω^P,A=ωA\widehat\omega_{P,A}=\omega_AωP,A​=ωA​ otherwise. The assertion is the conjunction of the following implication and converse implication: if C(A)<+∞C(A)<+\inftyC(A)<+∞, then qPn(A)>0q_{P_n}(A)>0qPn​​(A)>0 for all sufficiently large nnn; also, for all sufficiently large nnn, APnA_{P_n}APn​​ is nonempty and C(APn)<+∞C(A_{P_n})<+\inftyC(APn​​)<+∞; and

C(APn)⟶C(A),ω^Pn,A⟶ωA.C(A_{P_n})\longrightarrow C(A), \qquad \widehat\omega_{P_n,A}\longrightarrow\omega_A.C(APn​​)⟶C(A),ωPn​,A​⟶ωA​.

The capacity convergence uses the topology of [0,∞][0,\infty][0,∞], and the state convergence uses the state topology specified above. Conversely, if pPn(X)>0p_{P_n}(X)>0pPn​​(X)>0 for every n∈Nn\in\mathbb Nn∈N and every X∈AX\in AX∈A, and sup⁡n∈NC(APn)<+∞\sup_{n\in\mathbb N}C(A_{P_n})<+\inftysupn∈N​C(APn​​)<+∞, then C(A)<+∞C(A)<+\inftyC(A)<+∞. Here N\mathbb NN includes 000: the first implication gives eventual conclusions, whereas the positivity premise of the converse includes every index.

Human review
  • Endorsed by Shuze Chen · Oct 22, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 22, 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