Four opposite phase rotations: exact symmetric matrix average and polar form
ProvedQAlgorithms.Papers.Independent.Advance50.P42.fourPhase_matrix_averageLet be an arbitrary symmetric real matrix and let
Its average over the four phases is exactly
Equivalently this matrix is , where , , , and . With there exists such that and , giving the entire polar matrix in (157), including .
This identity describes the finite matrix average in Appendix B. For an arbitrary real symmetric matrix no uncertainty constraint or bounds are asserted. Applying it to the centered covariance of a mixed quantum state requires a separate means argument.
Formalization Note Coordinates are zero-based in Lean. The printed formula is replaced by existence of a polar angle, avoiding division by zero and the wrong quadrant. This is the pinned correction for the algebraic node; independent moderation is pending.
import Definitions.Def_QAlgorithms_Papers_Independent_Advance50_P42
noncomputable section
open scoped BigOperators Matrix
namespace QAlgorithms.Papers.Independent.Advance50.P42
/-- Appendix B, (153)--(157), PDF pp.20--21: the full four-phase real-matrix
identity, parameter identifications, and polar representation.
Source correction: the printed arctan(c1/c2) is replaced by existence of an angle,
which is the polar-representation claim used by (157), valid also when c2=0
or both mixed entries vanish. The arbitrary symmetric-matrix scope does not imply
these matrices are quantum covariances, nor the physical bounds a,b >= 1. -/
theorem fourPhase_matrix_average
(sigma : Matrix (Fin 4) (Fin 4) ℝ) (hsigma : sigma.IsSymm) :
let a := (sigma 0 0 + sigma 1 1) / 2
let b := (sigma 2 2 + sigma 3 3) / 2
let c1 := (sigma 0 3 + sigma 1 2) / 2
let c2 := (sigma 0 2 - sigma 1 3) / 2
let z := Real.sqrt (c1 ^ 2 + c2 ^ 2)
fourPhaseAverage sigma = (1 / 2 : ℝ) •
!![sigma 0 0 + sigma 1 1, 0, sigma 0 2 - sigma 1 3, sigma 0 3 + sigma 1 2;
0, sigma 0 0 + sigma 1 1, sigma 0 3 + sigma 1 2, -(sigma 0 2 - sigma 1 3);
sigma 0 2 - sigma 1 3, sigma 0 3 + sigma 1 2, sigma 2 2 + sigma 3 3, 0;
sigma 0 3 + sigma 1 2, -(sigma 0 2 - sigma 1 3), 0, sigma 2 2 + sigma 3 3] ∧
fourPhaseAverage sigma = phaseSymmetricMatrix a b c1 c2 ∧
∃ theta : ℝ, c1 = z * Real.sin theta ∧ c2 = z * Real.cos theta ∧
fourPhaseAverage sigma =
phaseSymmetricMatrix a b (z * Real.sin theta) (z * Real.cos theta) := by sorry
end QAlgorithms.Papers.Independent.Advance50.P42
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every real matrix satisfying , set
Define the notation
and, for real ,
Then all three of the following assertions hold:
and there exists a real number such that
The assertion includes the case , where and every real satisfies the stated conditions; uniqueness of is not asserted.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.