Degenerate optical Legendre transform and massive endpoint action
ProvedQAlgorithms.Papers.Independent.Renew46.P42.massiveAction_endpointLet and let , be the Gordon metrics of optical data , on an open spacetime patch . Consider a family , defined near for an open endpoint set , of affinely parametrized future timelike -geodesics with fixed initial endpoint , final endpoint for , and images in . Require and . Define
The endpoint function is differentiable and, with ,
For the massless Hamiltonian (34), for every nonzero . For a timelike laboratory tangent , the massive laboratory-time Lagrangian is
This constructs the principal function on extremals; the auxiliary mass is a regularization and does not assign a photon rest mass.
import Definitions.Def_QAlgorithms_Papers_Independent_Renew46_P42
namespace QAlgorithms.Papers.Independent.Renew46.P42
noncomputable section
open scoped BigOperators
open Filter Set MeasureTheory
-- Reuse the pinned accepted real quadratic form (arbitrary finite real matrices).
open QAlgorithms.Papers.Independent.Information100.P37 (quadratic)
/-- PDF pp.5–6, (39)–(49): zero massless Legendre transform and the massive
proper-length action on an actual local geodesic endpoint congruence.
The mass m is an artificial regularization, not a photon mass. -/
theorem massiveAction_endpoint (c m : ℝ) (P : OpticalPatch c)
(q : ℝ → V4 → V4) (z zBase : V4) (J : Set ℝ) (E : Set V4)
(hc : 0 < c) (hm : 0 < m) (hz : z ∈ E)
(hE : IsOpen E) (hEP : E ⊆ P.domain)
(hJ : IsOpen J) (h01 : Icc (0 : ℝ) 1 ⊆ J)
(hq : ContDiffOn ℝ 2 (fun a : ℝ × V4 => q a.1 a.2) (J ×ˢ E))
(hend : ∀ y ∈ E, q 0 y = zBase ∧ q 1 y = y)
(himage : ∀ y ∈ E, ∀ s ∈ Icc (0 : ℝ) 1, q s y ∈ P.domain)
(htime : ∀ y ∈ E, ∀ s ∈ Icc (0 : ℝ) 1,
0 < deriv (fun t => q t y) s 0)
(htimelike : ∀ y ∈ E, ∀ s ∈ Icc (0 : ℝ) 1,
0 < quadratic (patchDown P (q s y)) (deriv (fun t => q t y) s))
(hgeo : ∀ y ∈ E, ∀ s ∈ Icc (0 : ℝ) 1,
covAcceleration (patchDown P) (patchUp P) (fun t => q t y) s = 0) :
(∀ k : V3, k ≠ 0 →
dot3 k (rayVelocity c (P.index z) (P.flow z) k) -
rayHamiltonian c (P.index z) (P.flow z) k = 0) ∧
(∀ v : V3, 0 < quadratic (patchDown P z) ![c, v 0, v 1, v 2] →
massiveLagrangian c (P.index z) m (P.flow z) v =
-m * c * Real.sqrt (quadratic (patchDown P z) ![c, v 0, v 1, v 2])) ∧
DifferentiableAt ℝ (endpointAction P m q) z ∧
(∀ i : Fin 4, -cartesianPartial (endpointAction P m q) z i =
massiveCovector c m (patchDown P z) (deriv (fun s => q s z) 1) i) ∧
quadratic (patchUp P z) (gradient (endpointAction P m q) z) = m ^ 2 * c ^ 2 := by sorry
end
end QAlgorithms.Papers.Independent.Renew46.P42
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let , , , , and . Coordinate partial derivatives below are obtained by evaluating the Euclidean Fréchet derivative on the corresponding coordinate vector, and gradients collect these partial derivatives. For , put
Suppose , and consists of an open set and functions , , both on , such that and for . Write and . Define
Let , , and sets , satisfy: is open, ; is open and contains ; is jointly on ; and, for every , , , while for every ,
For real and , define
Define the function of the endpoint
Then every nonzero satisfies ; every with satisfies
is differentiable at ; and, writing , all four coordinate derivatives and their quadratic expression satisfy
Confirmed by the mission captain (proposal self-audit).
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.