Theorem 2, p. 3 — no nonzero jump-invariant subspace orthogonal to D ⇒ the only stationary states are the dark states
ProvedDissipativePrep.Unique.theorem_2Let be the Liouvillian of Eq. (2) with Hermitian , jump operators and rates . Fix a real number and assume that is a subspace in the sense of p. 3: every nonzero dark vector (annihilated by all and an eigenvector of ) has -eigenvalue , so . Suppose that there is no nonzero subspace with
Then every stationary state is a dark state: if is a density matrix with , then .
Together with the converse (dark states are stationary, §II) this characterizes all stationary states, and it is the uniqueness criterion applied in §IV.
Formalization Note The page quantifies over all subspaces ; the zero subspace is always invariant and orthogonal to , so the printed hypothesis would never hold, and the Lean statement quantifies over nonzero (P16's lazy). Rates are assumed. The standing assumption "all the states in correspond to the same eigenvalue of the Hamiltonian" is kept as a hypothesis.
import Mathlib import Definitions.Def_DissipativePrep_Unique_Model open scoped Matrix ComplexConjugate open QAlgorithms QAlgorithms.Papers.Independent.Dependency50.P16
namespace DissipativePrep.Unique
theorem theorem_2 {d m : ℕ} (H : Matrix (Fin d) (Fin d) ℂ) (hH : H.IsHermitian)
(c : Fin m → Matrix (Fin d) (Fin d) ℂ) (g : Fin m → ℝ) (hg : ∀ l, 0 < g l) (E : ℝ)
(hD : ∀ (ψ : EuclideanSpace ℂ (Fin d)) (E' : ℝ), ψ ≠ 0 → (∀ l, act (c l) ψ = 0) →
act H ψ = (E' : ℂ) • ψ → E' = E)
(hS : ∀ S : Submodule ℂ (EuclideanSpace ℂ (Fin d)), lazy c S → ¬ S ≤ (darkSpace H c E)ᗮ) :
∀ ρ : Matrix (Fin d) (Fin d) ℂ, stationary (krausData H hH c g) ρ →
densitySupport ρ ≤ darkSpace H c E := by sorry
end DissipativePrep.Unique
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.