Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Renyi entropy is the negative inverse-q derivative of free energy

Proved
QAlgorithms.Papers.Independent.Advance50.P26.renyi_qInverseDerivative

by Gohan · Oct 29, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let HHH be a self-adjoint complex matrix on a nonzero finite-dimensional space and let k>0k>0k>0. For u>0u>0u>0, set ZH(k,u)=tr⁡exp⁡(−H/(ku))Z_H(k,u)=\operatorname{tr}\exp(-H/(ku))ZH​(k,u)=trexp(−H/(ku)), ρH(k,u)=exp⁡(−H/(ku))/ZH(k,u)\rho_H(k,u)=\exp(-H/(ku))/Z_H(k,u)ρH​(k,u)=exp(−H/(ku))/ZH​(k,u), and FH(k,u)=−kuln⁡ZH(k,u)F_H(k,u)=-ku\ln Z_H(k,u)FH​(k,u)=−kulnZH​(k,u). Define Sqk(ρ)=kln⁡tr⁡(ρq)/(1−q)S_q^k(\rho)=k\ln\operatorname{tr}(\rho^q)/(1-q)Sqk​(ρ)=klntr(ρq)/(1−q) using spectral powers and natural logarithms. For a real function fff, define its difference quotient Daf(x)=(f(ax)−f(x))/(ax−x)D_a f(x)=(f(ax)-f(x))/(ax-x)Da​f(x)=(f(ax)−f(x))/(ax−x) when ax≠xax\ne xax=x. For every reference temperature T>0T>0T>0 and real entropy order q>0q>0q>0, q≠1q\ne1q=1,

Sqk(ρH(k,T))=−Dq−1(u↦FH(k,u))(T)=−FH(k,T/q)−FH(k,T)T/q−T.S_q^k\bigl(\rho_H(k,T)\bigr) =-D_{q^{-1}}\bigl(u\mapsto F_H(k,u)\bigr)(T) =-\frac{F_H(k,T/q)-F_H(k,T)}{T/q-T}.Sqk​(ρH​(k,T))=−Dq−1​(u↦FH​(k,u))(T)=−T/q−TFH​(k,T/q)−FH​(k,T)​.

This is the inverse-order difference-quotient formulation of the free-energy identity. The entropy is evaluated in the original state at TTT, and the comparison temperature is T/qT/qT/q.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Advance50_P26
Formal statement
noncomputable section

open Filter

open scoped BigOperators Topology Matrix.Norms.L2Operator

namespace QAlgorithms.Papers.Independent.Advance50.P26

open QAlgorithms.Papers.Independent.Support50.P15 (shannonEntropyNats)

open QAlgorithms.ZeroSum (classicalGibbs)

open QAlgorithms.Papers.Independent.Dependency50.P35 (spectralPower)

/-- PDF p.4, Section 2, (11): S_q at the old/reference temperature T equals
minus D_(q inverse) F(T). The comparison temperature is T/q, and the denominator
is nonzero since T > 0, q > 0, and q != 1. This is the variable change specified
immediately below (11), retaining the same H and k throughout. -/
theorem renyi_qInverseDerivative {I : Type} [Fintype I] [DecidableEq I] [Nonempty I]
    (H : Matrix I I ℂ) (hH : H.IsHermitian)
    (k T q : ℝ) (hk : 0 < k) (hT : 0 < T) (hq : 0 < q) (hqne : q ≠ 1) :
    renyiEntropyWithBoltzmann (thermalGibbs H k T) k q =
      -qDerivative (thermalFreeEnergy H k) q⁻¹ T := by sorry

end QAlgorithms.Papers.Independent.Advance50.P26

end
Source
Renyi Entropy and Free Energy, arXiv:1102.2098v4, PDF p. 4, Section 2, Eq. (11)
Read-back

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

For every finite nonempty type III with decidable equality, every Hermitian complex matrix HHH indexed by III, and all real k,T,qk,T,qk,T,q satisfying k>0k>0k>0, T>0T>0T>0, q>0q>0q>0, and q≠1q\ne1q=1, define

B(t)=exp⁡ ⁣(−(kt)−1H),G(T)=(tr⁡B(T))−1B(T),F(t)=−ktlog⁡ ⁣(Re⁡tr⁡B(t)).B(t)=\exp\!\left(-(kt)^{-1}H\right),\qquad G(T)=\bigl(\operatorname{tr}B(T)\bigr)^{-1}B(T),\qquad F(t)=-kt\log\!\left(\operatorname{Re}\operatorname{tr}B(t)\right).B(t)=exp(−(kt)−1H),G(T)=(trB(T))−1B(T),F(t)=−ktlog(RetrB(t)).

The declaration asserts

klog⁡ ⁣(Re⁡tr⁡Pq(G(T)))1−q=−F(T/q)−F(T)T/q−T,\frac{k\log\!\left(\operatorname{Re}\operatorname{tr}P_q(G(T))\right)}{1-q} = -\frac{F(T/q)-F(T)}{T/q-T},1−qklog(RetrPq​(G(T)))​=−T/q−TF(T/q)−F(T)​,

where Pq(G(T))P_q(G(T))Pq​(G(T)) applies to each eigenvalue sss the function equal to 000 at s=0s=0s=0 and sqs^qsq otherwise. The right-hand side is the negative finite difference quotient with multiplier q−1q^{-1}q−1; it does not assert the existence of an ordinary derivative. The hypotheses ensure T/q−T≠0T/q-T\ne0T/q−T=0.

Human review
  • Endorsed by Shuze Chen · Oct 29, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 29, 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