Relabelling the basis of a POVM along a bijection of index sets
ProvedBosonicLearn.fockPOVM_reindexA POVM can be transported along a relabelling of the basis.
Let be a POVM on with outcomes in a measurable space and let be a bijection of index sets. Then there is a POVM on with outcomes in such that for every density operator on the outcome law of on the relabelled operator
equals the outcome law of on .
Role. The -copy Fock space has a product basis that can be indexed by -tuples of Fock labels or by nested pairs. This statement passes between the two, as needed when a protocol is built round by round in BosonicLearn.adaptive_scheme_product_povm.
Formalization Note Both POVMs are FockPOVMs, and density operators are IsDensity of Dependency50.P33. The index sets are arbitrary.
Related platform material The space used here (Hilbert A) is the same Lean type as OAI.TensorLp.H A, and is the same type as OAI.EntropyPhotonNumber.Fock n. The published definitions OAI_Analysis_TensorLp (tensor products, slices and operator lifts on ) and OAI_MathematicalPhysics_Fock_Operators (OAI.CrossEnsemble.operator, the bounded operator of a family, with hasSum_operator and entry_hasSum) are sorry-free and can be imported. Proved lemmas of this mission that can be imported: BosonicLearn.traceDist_le_one (nuclear expansions, truncation to finite index sets) BosonicLearn.effect_matrix_operator (effect matrices as bounded positive operators), BosonicLearn.effect_trace_pairing (Born sums along nuclear expansions), BosonicLearn.fockPOVM_form_measure (effects are weakly -additive) and BosonicLearn.fockPOVM_postprocess.
import Mathlib import Definitions.Def_BosonicLearn_FockPOVM open QAlgorithms.Papers.Independent.Dependency50.P33 (IsDensity) open MeasureTheory open scoped ENNReal Matrix
namespace BosonicLearn
/-- **Relabelling the basis of a POVM.**
Chen, Anna Mele, Anna Mele, Preskill, *When are bosonic Gaussian states classical to learn?*, arXiv:2609.26705, §1.1, p. 4: "A natural basis for H^{(n)} is the Fock basis {|m⟩ ≡ |m_1⟩ ⊗ ⋯ ⊗ |m_n⟩}_{m ∈ ℕⁿ}"; the
`N`-copy space has the product basis, which can be indexed by tuples or by nested pairs.
Let `M` be a POVM on `ℓ²(J)` with outcomes in `Ω` and let `e : J ≃ J'` be a bijection of index sets. There is a
POVM `M'` on `ℓ²(J')` with outcomes in `Ω` such that for every density operator `ρ` on `ℓ²(J)` the outcome law of
`M'` on the relabelled operator `ρ'(i, j) = ρ(e⁻¹ i, e⁻¹ j)` is the outcome law of `M` on `ρ`.
**Formalization Note.** Both are `FockPOVM`s. The statement is used to pass between the index types
`(Fin k → ℕⁿ) × ℕⁿ` and `Fin (k + 1) → ℕⁿ` of `k + 1` copies. -/
theorem fockPOVM_reindex {J J' Ω : Type} [DecidableEq J] [DecidableEq J'] [MeasurableSpace Ω]
(M : FockPOVM J Ω) (e : J ≃ J') :
∃ M' : FockPOVM J' Ω, ∀ ρ : Matrix J J ℂ, IsDensity ρ →
M'.law (fun i j => ρ (e.symm i) (e.symm j)) = M.law ρ := by sorry
end BosonicLearn