Displacing two states by the same vector: and the trace distance is unchanged
ProvedBosonicLearn.state_displace_pairbosonic-gaussian-statesquantum-informationquantum-learning
Displaced states and invariance of the trace norm under a common displacement.
Let be states on the -mode Fock space and . Then there are states — the displaced states and — such that
where .
Role. In the proof of Theorem 8 the estimated mean is corrected by a displacement, which "does not change their trace distance"; with BosonicLearn.state_ext_characteristic this gives BosonicLearn.traceDist_gaussian_common_mean.
Formalization Note and are in the interleaved quadrature coordinates of Support50.P14.oscillatorCharacteristic, is meanForm δ ξ, and the trace norm is traceSize.
Preamble
import Mathlib import Definitions.Def_BosonicLearn_husimiQ import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P14 open QAlgorithms.Papers.Independent.Dependency50.P33 (State traceSize) open QAlgorithms.Papers.Independent.Claimed50.P47 (Occupation) open QAlgorithms.Papers.Independent.Support50.P14 open MeasureTheory open scoped ENNReal
Formal statement
namespace BosonicLearn
/-- **Displacing two states by the same phase-space vector.**
Chen, Anna Mele, Anna Mele, Preskill, *When are bosonic Gaussian states classical to learn?*, arXiv:2609.26705, proof of Theorem 8, p. 62: "a common displacement is implemented by the same unitary on both states, so it does not change
their trace distance"; §1.1, pp. 4–5 (displacement operators).
For states `ρ, σ` on the `n`-mode Fock space and `δ ∈ ℝ²ⁿ` there are states `ρ', σ'` (the displaced states
`D_δ ρ D_δ†`, `D_δ σ D_δ†`) with characteristic functions `χ_{ρ'}(ξ) = e^{iξ·δ} χ_ρ(ξ)`, `χ_{σ'}(ξ) = e^{iξ·δ} χ_σ(ξ)`,
and `‖ρ' − σ'‖₁ = ‖ρ − σ‖₁`.
**Formalization Note.** `δ` and `ξ` are in the interleaved quadrature coordinates of
`Support50.P14.oscillatorCharacteristic`, and `ξ·δ` is `meanForm δ ξ`. The trace norm is `traceSize`. -/
theorem state_displace_pair {n : ℕ} (ρ σ : State (Occupation (Fin n))) (δ : Fin n × Fin 2 → ℝ) :
∃ ρ' σ' : State (Occupation (Fin n)),
(∀ ξ : Fin n × Fin 2 → ℝ, oscillatorCharacteristic ρ'.coefficients ξ =
Complex.exp (Complex.I * (meanForm δ ξ : ℂ)) * oscillatorCharacteristic ρ.coefficients ξ) ∧
(∀ ξ : Fin n × Fin 2 → ℝ, oscillatorCharacteristic σ'.coefficients ξ =
Complex.exp (Complex.I * (meanForm δ ξ : ℂ)) * oscillatorCharacteristic σ.coefficients ξ) ∧
traceSize (ρ'.coefficients - σ'.coefficients) = traceSize (ρ.coefficients - σ.coefficients) := by sorry
end BosonicLearn
Source
Chen, Anna Mele, Anna Mele, Preskill, When are bosonic Gaussian states classical to learn?, arXiv:2609.26705, proof of Theorem 8, p. 62 ("a common displacement is implemented by the same unitary on both states, so it does not change their trace distance"; the estimator's covariance dominates the true one)