is the Gaussian state with mean and covariance (real form)
ProvedBosonicLearn.fugacity_isGaussianStateThe fugacity form is a Gaussian state, with explicit covariance.
Let be a positive semidefinite complex matrix with positive definite, and (Hermitian). Then is the coefficient array of a state , and is the Gaussian state with
in the grouped order .
Checks. gives the vacuum . For one mode, , the thermal state with mean photon number . The formula was also checked numerically against the definition for a random -mode .
Role. Explicit form of clause 1 of Lemma 2 of the paper. It gives directly that is passive (with BosonicLearn.gammaFock_basic) and that implies (BosonicLearn.fugacity_promiseClass).
Formalization Note IsGaussianState is the mission predicate (characteristic function, finite energy, symmetric covariance, first and second moments); the covariance is written with Matrix.fromBlocks and entrywise real and imaginary parts of .
import Mathlib import Definitions.Def_BosonicLearn_fugacityCoeff import Definitions.Def_BosonicLearn_IsGaussianState open QAlgorithms.Papers.Independent.Dependency50.P33 (State) open QAlgorithms.Papers.Independent.Claimed50.P47 (Occupation) open scoped Matrix ComplexOrder
namespace BosonicLearn
/-- **`det(I − R)Γ(R)` is the Gaussian state with mean `0` and covariance `½ (1 + R)(1 − R)⁻¹`.**
Chen, Anna Mele, Anna Mele, Preskill, *When are bosonic Gaussian states classical to learn?*, arXiv:2609.26705, Lemma 2, Eq. (21), p. 20: "Given R, which we call the fugacity matrix, the corresponding passive Gaussian state is
ρ = det(I − R)Γ(R)."
Let `R` be positive semidefinite with `1 − R` positive definite, and `C = (1 + R)(1 − R)⁻¹` (Hermitian). Then
`det(1 − R)Γ(R)` is the coefficient array of a state, and that state is Gaussian with mean `0` and covariance
matrix, in the grouped order `(q₁,…,qₙ,p₁,…,pₙ)`,
`Σ = ½ [[Re C, −Im C], [Im C, Re C]]`.
**Formalization Note.** Explicit form of the first clause of `lemma_2`. For `R = 0` this is the vacuum, `Σ = ½ I`;
for one mode, `Σ = ½ (1 + d)/(1 − d) · I`, the thermal state with mean photon number `d/(1 − d)`. -/
theorem fugacity_isGaussianState {n : ℕ} (R : Matrix (Fin n) (Fin n) ℂ) (hR : R.PosSemidef)
(hR1 : (1 - R).PosDef) :
∃ ρ : State (Occupation (Fin n)), ρ.coefficients = fugacityCoeff R ∧
IsGaussianState ρ 0 ((1 / 2 : ℝ) • Matrix.fromBlocks
(((1 + R) * (1 - R)⁻¹).map Complex.re) (-(((1 + R) * (1 - R)⁻¹).map Complex.im))
(((1 + R) * (1 - R)⁻¹).map Complex.im) (((1 + R) * (1 - R)⁻¹).map Complex.re)) := by sorry
end BosonicLearn