Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

The explicit function separates the standard family member

Proved
QAlgorithms.Papers.Independent.Intake100.P66.separatingF_separates

by Gohan · Oct 30, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Fix an integer n≥2n\ge2n≥2, let N1={x1,…,xn}N_1=\{x_1,\ldots,x_n\}N1​={x1​,…,xn​} and N2={a,b,c}N_2=\{a,b,c\}N2​={a,b,c} be disjoint, and set N=N1⊔N2N=N_1\sqcup N_2N=N1​⊔N2​. For J⊆N1J\subseteq N_1J⊆N1​ and K⊆N2K\subseteq N_2K⊆N2​, define

f(JK)=θK+∣J∣λK−μJK.f(JK)=\theta_K+|J|\lambda_K-\mu_{JK}.f(JK)=θK​+∣J∣λK​−μJK​.

In the column order K=(∅,a,b,c,ab,ac,bc,abc)K=(\varnothing,a,b,c,ab,ac,bc,abc)K=(∅,a,b,c,ab,ac,bc,abc), the tables are

θn+1=(0,2n3+5n2,4n2+2n−1,3n2+n,2n3+8n2+4n−1,2n3+5n2+2n,6n2+4n−1,2n3+8n2+4n−1),\frac{\theta}{n+1}=\bigl(0,2n^3+5n^2,4n^2+2n-1,3n^2+n,2n^3+8n^2+4n-1,2n^3+5n^2+2n,6n^2+4n-1,2n^3+8n^2+4n-1\bigr),n+1θ​=(0,2n3+5n2,4n2+2n−1,3n2+n,2n3+8n2+4n−1,2n3+5n2+2n,6n2+4n−1,2n3+8n2+4n−1), −λ=(n2,2n3+5n2+2n,4n2+2n−1,4n2+2n−1,2n3+8n2+4n−1,2n3+5n2+2n,6n2+4n−1,2n3+8n2+4n−1).-\lambda=\bigl(n^2,2n^3+5n^2+2n,4n^2+2n-1,4n^2+2n-1,2n^3+8n^2+4n-1,2n^3+5n^2+2n,6n^2+4n-1,2n^3+8n^2+4n-1\bigr).−λ=(n2,2n3+5n2+2n,4n2+2n−1,4n2+2n−1,2n3+8n2+4n−1,2n3+5n2+2n,6n2+4n−1,2n3+8n2+4n−1).

The only nonzero μ\muμ values are μ{a}=2n2(n+1)\mu_{\{a\}}=2n^2(n+1)μ{a}​=2n2(n+1) and μ{a,b}=2n(n+1)2\mu_{\{a,b\}}=2n(n+1)^2μ{a,b}​=2n(n+1)2, on those exact subsets of NNN.

For pairwise disjoint subsets U,V,W,α1,…,αpU,V,W,\alpha_1,\ldots,\alpha_pU,V,W,α1​,…,αp​ of a fixed finite ground set, empty subsets permitted, put

Lp(h;U,V,W,α)=∑i=1pIh(U:V∣αi)+∑i=1ph(αi)−h ⁣(⋃i=1pαi)−(p−1)Ih(UV:W).L_p(h;U,V,W,\alpha)=\sum_{i=1}^p I_h(U:V\mid\alpha_i)+\sum_{i=1}^p h(\alpha_i)-h\!\left(\bigcup_{i=1}^p\alpha_i\right)-(p-1)I_h(UV:W).Lp​(h;U,V,W,α)=i=1∑p​Ih​(U:V∣αi​)+i=1∑p​h(αi​)−h(i=1⋃p​αi​)−(p−1)Ih​(UV:W).

Here hhh is a normalized real set function, Ih(U:V)=h(U)+h(V)−h(UV)I_h(U:V)=h(U)+h(V)-h(UV)Ih​(U:V)=h(U)+h(V)−h(UV) and Ih(U:V∣W)=h(UW)+h(VW)−h(W)−h(UVW)I_h(U:V\mid W)=h(UW)+h(VW)-h(W)-h(UVW)Ih​(U:V∣W)=h(UW)+h(VW)−h(W)−h(UVW); juxtaposition denotes union. The constrained inequality CpC_pCp​ is Lp≥0L_p\ge0Lp​≥0 under Ih(U:W∣V)=Ih(V:W∣U)=0I_h(U:W\mid V)=I_h(V:W\mid U)=0Ih​(U:W∣V)=Ih​(V:W∣U)=0.

The function satisfies If(a:c∣b)=If(b:c∣a)=0I_f(a:c\mid b)=I_f(b:c\mid a)=0If​(a:c∣b)=If​(b:c∣a)=0. For every integer p≥2p\ge2p≥2 with p≠np\ne np=n and every pairwise disjoint family αi⊆N1\alpha_i\subseteq N_1αi​⊆N1​,

Lp(f;a,b,c,α)≥0,Ln(f;a,b,c,({xi})i=1n)<0.L_p(f;a,b,c,\alpha)\ge0,\qquad L_n(f;a,b,c,(\{x_i\})_{i=1}^n)<0.Lp​(f;a,b,c,α)≥0,Ln​(f;a,b,c,({xi​})i=1n​)<0.

The labels a,b,ca,b,ca,b,c are fixed in the first assertion and the αi\alpha_iαi​ may be empty. This is a separation by the explicit table, without requiring fff to obey weak monotonicity.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intake100_P66
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intake100.P66

open scoped BigOperators

open QAlgorithms (IsDensityMatrix vnEntropy)

attribute [local instance] Classical.propDecidable Classical.decEq

/-- PDF pp.8--9, Proposition 4: fixed a,b,c and disjoint X-subsets, including empty
ones; all other p>=2 families hold, but the standard C_n residual is negative. -/
theorem separatingF_separates (n : ℕ) (hn : 2 ≤ n) :
    evaluateEntropyForm (cmiForm {entropyLabelA} {c} {b}) (separatingF n) = 0 ∧
    evaluateEntropyForm (cmiForm {b} {c} {entropyLabelA}) (separatingF n) = 0 ∧
    (∀ (p : ℕ), 2 ≤ p → p ≠ n → ∀ alpha : Fin p → Finset (entropyParty n),
      (∀ i, alpha i ⊆ Finset.univ.filter (fun j => ∃ t : Fin n, j = x t)) →
      (∀ i j, i ≠ j → Disjoint (alpha i) (alpha j)) →
      0 ≤ evaluateEntropyForm (familyForm {entropyLabelA} {b} {c} alpha) (separatingF n)) ∧
    evaluateEntropyForm (standardFamily n) (separatingF n) < 0 := by sorry

end QAlgorithms.Papers.Independent.Intake100.P66

end
Source
Infinitely many constrained inequalities for the von Neumann entropy, arXiv:1107.0624v1, PDF p. 8, Proposition 4, §IV: the explicit f separates C_n from the other C_p
Read-back

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

For every natural number n≥2n\ge2n≥2, let Pn={A,B,C}⊔{Xi:0≤i<n}P_n=\{A,B,C\}\sqcup\{X_i:0\le i<n\}Pn​={A,B,C}⊔{Xi​:0≤i<n}, with distinct labels, and set X={Xi:0≤i<n}X=\{X_i:0\le i<n\}X={Xi​:0≤i<n}. For M⊆PnM\subseteq P_nM⊆Pn​, define K=M∩{A,B,C}K=M\cap\{A,B,C\}K=M∩{A,B,C}, r=∣M∩X∣r=|M\cap X|r=∣M∩X∣, and

Fn(M)=(n+1)Tn(K)−rLn(K)−2n2(n+1)1M={A}−2n(n+1)21M={A,B},F_n(M)=(n+1)T_n(K)-rL_n(K) -2n^2(n+1)\mathbf1_{M=\{A\}} -2n(n+1)^2\mathbf1_{M=\{A,B\}},Fn​(M)=(n+1)Tn​(K)−rLn​(K)−2n2(n+1)1M={A}​−2n(n+1)21M={A,B}​,

where 1\mathbf11 denotes an indicator and

KTn(K)Ln(K)∅0n2{A}2n3+5n22n3+5n2+2n{B}4n2+2n−14n2+2n−1{C}3n2+n4n2+2n−1{A,B} or {A,B,C}2n3+8n2+4n−12n3+8n2+4n−1{A,C}2n3+5n2+2n2n3+5n2+2n{B,C}6n2+4n−16n2+4n−1\begin{array}{c|c|c} K&T_n(K)&L_n(K)\\ \hline \varnothing&0&n^2\\ \{A\}&2n^3+5n^2&2n^3+5n^2+2n\\ \{B\}&4n^2+2n-1&4n^2+2n-1\\ \{C\}&3n^2+n&4n^2+2n-1\\ \{A,B\}\text{ or }\{A,B,C\}&2n^3+8n^2+4n-1&2n^3+8n^2+4n-1\\ \{A,C\}&2n^3+5n^2+2n&2n^3+5n^2+2n\\ \{B,C\}&6n^2+4n-1&6n^2+4n-1 \end{array}K∅{A}{B}{C}{A,B} or {A,B,C}{A,C}{B,C}​Tn​(K)02n3+5n24n2+2n−13n2+n2n3+8n2+4n−12n3+5n2+2n6n2+4n−1​Ln​(K)n22n3+5n2+2n4n2+2n−14n2+2n−12n3+8n2+4n−12n3+5n2+2n6n2+4n−1​​

For subsets of PnP_nPn​, write

JF(U,V;W)=Fn(U∪W)+Fn(V∪W)−Fn(W)−Fn(U∪V∪W),IF(U;V)=Fn(U)+Fn(V)−Fn(U∪V),J_F(U,V;W)=F_n(U\cup W)+F_n(V\cup W)-F_n(W)-F_n(U\cup V\cup W), \quad I_F(U;V)=F_n(U)+F_n(V)-F_n(U\cup V),JF​(U,V;W)=Fn​(U∪W)+Fn​(V∪W)−Fn​(W)−Fn​(U∪V∪W),IF​(U;V)=Fn​(U)+Fn​(V)−Fn​(U∪V),

using a party label as shorthand for its singleton. The assertion is the conjunction of

JF(A,C;B)=0,JF(B,C;A)=0,J_F(A,C;B)=0,\qquad J_F(B,C;A)=0,JF​(A,C;B)=0,JF​(B,C;A)=0,

the statement that for every natural number p≥2p\ge2p≥2 with p≠np\ne np=n and every indexed collection α0,…,αp−1\alpha_0,\ldots,\alpha_{p-1}α0​,…,αp−1​ of pairwise disjoint subsets of XXX,

0≤∑i=0p−1JF(A,B;αi)+∑i=0p−1Fn(αi)−Fn ⁣(⋃i=0p−1αi)−(p−1)IF(AB;C),0\le \sum_{i=0}^{p-1}J_F(A,B;\alpha_i) +\sum_{i=0}^{p-1}F_n(\alpha_i) -F_n\!\left(\bigcup_{i=0}^{p-1}\alpha_i\right) -(p-1)I_F(AB;C),0≤i=0∑p−1​JF​(A,B;αi​)+i=0∑p−1​Fn​(αi​)−Fn​(i=0⋃p−1​αi​)−(p−1)IF​(AB;C),

and the strict inequality

∑i=0n−1JF(A,B;Xi)+∑i=0n−1Fn(Xi)−Fn(X)−(n−1)IF(AB;C)<0.\sum_{i=0}^{n-1}J_F(A,B;X_i) +\sum_{i=0}^{n-1}F_n(X_i)-F_n(X) -(n-1)I_F(AB;C)<0.i=0∑n−1​JF​(A,B;Xi​)+i=0∑n−1​Fn​(Xi​)−Fn​(X)−(n−1)IF​(AB;C)<0.

Here AB={A,B}AB=\{A,B\}AB={A,B}; the sets αi\alpha_iαi​ may be empty, need not cover XXX, and ppp may exceed nnn.

Human review
  • Endorsed by Shuze Chen · Oct 30, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 30, 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