Submultiplicativity of the dilatation `K (f ∘ g) ≤ K f * K g`.
ProvedTeichmuller.LinMap.dil_comp_leaether-cataloggeometry
Submultiplicativity of the dilatation K (f ∘ g) ≤ K f * K g.
theorem Teichmuller.LinMap.dil_comp_le: (f.comp g).dil ≤ f.dil * g.dil := by sorry
Formalization Note Transplanted verbatim from the Aether Catalog source Geometry/Teichmuller/LinearQC.lean; the statement is byte-identical to the source declaration, elaborated with autoImplicit disabled in the platform environment.
Preamble
-- Thm stub generated from Geometry/Teichmuller/LinearQC.lean
import Mathlib
import Definitions.Def_Geometry_Teichmuller_LinearQC
/-
# Quasiconformal dilatation of real-linear maps of the plane
This file develops, from scratch, the *infinitesimal* layer of Teichmüller theory: the
quasiconformal dilatation of an orientation-preserving, nonsingular `ℝ`-linear map of `ℂ`.
Every `ℝ`-linear map `f : ℂ → ℂ` has the unique normal form
f z = a * z + b * conj z , a b : ℂ,
with Jacobian determinant `‖a‖² - ‖b‖²`; the map is orientation preserving and invertible
exactly when `‖b‖ < ‖a‖`. Its *Beltrami coefficient* is `μ = b / a` and its *dilatation* is
K f = (‖a‖ + ‖b‖) / (‖a‖ - ‖b‖) = (1 + ‖μ‖) / (1 - ‖μ‖),
the ratio of the major to the minor axis of the ellipse `f '' (unit circle)`.
The mathematical core of the file is the exact **determinant identity**
‖A‖² - ‖B‖² = (‖a‖² - ‖b‖²) * (‖c‖² - ‖d‖²)
for the composite `(A, B)` of `(a, b)` and `(c, d)` (`LinMap.normSq_sub_normSq_comp`), which
combined with the crude triangle bound `‖A‖ + ‖B‖ ≤ (‖a‖ + ‖b‖)(‖c‖ + ‖d‖)` yields
**submultiplicativity of the dilatation** `K (f ∘ g) ≤ K f * K g` through the identity
`K = (‖a‖ + ‖b‖)² / (‖a‖² - ‖b‖²)`. This is the statement that makes `log K` a metric
downstream (`Geometry.Teichmuller.TorusSpace`).
Main results:
* `LinMap.one_le_dil` : dilatation is at least `1`;
* `LinMap.dil_eq_one_iff` : dilatation `1` ⇔ the map is conformal (`ℂ`-linear);
* `LinMap.comp_apply` : the composition formula is correct as maps of `ℂ`;
* `LinMap.dil_comp_le` : `K (f ∘ g) ≤ K f * K g`;
* `LinMap.dil_inv` : `K (f⁻¹) = K f`, and `LinMap.inv_apply` verifies the inverse formula;
* `LinMap.dil_eq_beltrami` : `K = (1 + ‖μ‖) / (1 - ‖μ‖)`.
-- !-- Lab Notes -- !--
Hypothesizer: the submultiplicativity `K(f∘g) ≤ K(f)K(g)` — the axiom that makes the
Teichmüller metric a metric — should be a purely algebraic identity, not an analytic estimate.
Experimenter: the naive bound `‖A‖ - ‖B‖ ≥ (‖a‖-‖b‖)(‖c‖-‖d‖)` is FALSE termwise (the triangle
inequality loses `2‖b‖‖d‖`); the repair is to bound the *product* `(‖A‖-‖B‖)(‖A‖+‖B‖)` exactly:
the cross terms `2Re(a c conj b conj d)` in `‖A‖²` and `‖B‖²` cancel identically, giving the
determinant identity. Analyst: the lost factor is exactly the multiplicativity of the Jacobian;
the correct formulation of `K` for the estimate is `K = (‖a‖+‖b‖)²/(‖a‖²-‖b‖²)`, i.e. "square of
an operator norm over a determinant", which is the shape that composes.
-/
open Teichmuller
open Complex
open LinMap
variable (f g : LinMap)Formal statement
theorem Teichmuller.LinMap.dil_comp_le: (f.comp g).dil ≤ f.dil * g.dil := by sorry
Source