Renyi entropy is the negative temperature secant slope of free energy
ProvedQAlgorithms.Papers.Independent.Advance50.P26.renyi_freeEnergyLet be a self-adjoint complex matrix on a nonzero finite-dimensional space, and let be Boltzmann's constant. For each positive temperature , define
For a density matrix and a real order , , define its Renyi entropy by , where the power is spectral and all logarithms are natural. For any positive temperatures with , one has
This relates entropy at the original temperature to an equilibrium free-energy difference for the same Hamiltonian and Boltzmann constant. No positivity condition on or zero-reference-free-energy condition is required.
import Definitions.Def_QAlgorithms_Papers_Independent_Advance50_P26
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.3, Section 2, (8)-(9): the entropy of the Gibbs state at T0 is the
negative secant slope of free energy between T0 and T. Here q is constructed as T0/T;
positive distinct temperatures ensure q > 0 and q != 1. No hypothesis F(T0) = 0 or
H positive semidefinite is imposed. The p.3 source corrections are documented above. -/
theorem renyi_freeEnergy {I : Type} [Fintype I] [DecidableEq I] [Nonempty I]
(H : Matrix I I ℂ) (hH : H.IsHermitian)
(k T0 T : ℝ) (hk : 0 < k) (hT0 : 0 < T0) (hT : 0 < T) (hneq : T ≠ T0) :
renyiEntropyWithBoltzmann (thermalGibbs H k T0) k (T0 / T) =
-(thermalFreeEnergy H k T - thermalFreeEnergy H k T0) / (T - T0) := by sorry
end QAlgorithms.Papers.Independent.Advance50.P26
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every finite nonempty type with decidable equality, every Hermitian complex matrix indexed by , and all real satisfying , , , and , define
The declaration asserts
where applies to each eigenvalue the function that is at and otherwise. Here and , so the entropy definition uses its branch for orders different from , and both displayed denominators are nonzero.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.