Hong–Ou–Mandel output in the paper’s real convention
ProvedQAlgorithms.Papers.Independent.Frontier18.P14.hongOuMandelLet . For one photon in each matched input mode, the full passive transformation gives
This vector is normalized. Joint photon counting gives probability to each of and and zero to every other outcome, including coincidence . The relative minus sign fixes the source's reflection convention.
import Definitions.Def_QAlgorithms_Papers_Independent_Frontier18_P14
noncomputable section
open scoped Matrix ComplexConjugate
open Filter
attribute [local instance] Classical.propDecidable
namespace QAlgorithms.Papers.Independent.Frontier18.P14
open QAlgorithms.Papers.Independent.Dependency50.P34 (fockKet)
/-- PDF pp.22–23, (4.23)–(4.24): the real-reflection Hong–Ou–Mandel
output with its relative minus sign and no coincidence counts. -/
theorem hongOuMandel :
passiveApply balanced (fockKet 1 1) =
(1 / (Real.sqrt 2 : ℂ)) • (fockKet 2 0 - fockKet 0 2) ∧
‖passiveApply balanced (fockKet 1 1)‖ = 1 ∧
(∀ mn : ℕ × ℕ, Complex.normSq (passiveApply balanced (fockKet 1 1) mn) =
if mn = (2, 0) ∨ mn = (0, 2) then (1 / 2 : ℝ) else 0) := by sorry
end QAlgorithms.Papers.Independent.Frontier18.P14
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let , with coordinate basis vectors , and let
The defined passive action of on gives the finite vector
the definition forms output coefficients by summing over input occupations with the same total occupation and uses zero as a fallback if the resulting coefficient array is not square summable, a fallback that does not occur for this finite array. The theorem asserts
and, for every ,
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.