K24: actual scalar and matrix inverse remainders in real-order Sobolev norms
ProvedK24.Global.realOrderSobolevInverseRemaindersFor every real r>2, sufficiently small continuous scalar perturbations u in the original local H^r norm have actual pointwise inverse differences (1+u)^(-1)-1 in that same Sobolev class, with norm at most twice the perturbation norm and an original-norm quadratic bound for the remainder (1+u)^(-1)-1+u. The same statement holds for actual 4 by 4 real matrix functions, with the unchanged matrixSobolevENN square-sum norm, actual determinant nonvanishing, actual matrix inverse, and entrywise continuous H^r representatives. Thresholds and finite remainder constants are chosen before the perturbation and depend only on r. No identity/constant function is asserted to belong to L2 on R4. The proof constructs the complete continuous Sobolev carrier with the exact original norm and applies a genuine contraction v -> -u-uv. Its fixed-point identity proves pointwise invertibility and identifies v with the actual inverse difference; v+u=-uv gives the quadratic bound. This is derived inverse-coefficient analytic infrastructure for the original Sobolev branch. It does not assert the global manifold partialGauge Frechet field or close the uniform analytic core. Parent https://prove2.me/theorems/06fdeaf7-3709-47fc-b5c6-674da98954a4 .
import Definitions.Def_K24GlobalStatement set_option autoImplicit false universe u noncomputable section open scoped NNReal ENNReal Topology Manifold ContDiff set_option backward.isDefEq.respectTransparency false set_option maxRecDepth 10000 set_option maxHeartbeats 1200000 set_option synthInstance.maxHeartbeats 200000 set_option maxSynthPendingDepth 16
theorem K24.Global.realOrderSobolevInverseRemainders :
( ∀ (r : ℝ), 2 < r → ∃ δ : ℝ, 0 < δ ∧ ∃ C : ℝ≥0,
∀ u : K24.Global.SobolevModel → ℝ,
Continuous u → K24.Global.LocalSobolevMem r u →
K24.Global.localSobolevENN r u < ENNReal.ofReal δ →
(∀ x, 1 + u x ≠ 0) ∧
Continuous (fun x => (1 + u x)⁻¹ - 1) ∧
K24.Global.LocalSobolevMem r (fun x => (1 + u x)⁻¹ - 1) ∧
K24.Global.localSobolevENN r (fun x => (1 + u x)⁻¹ - 1) ≤
2 * K24.Global.localSobolevENN r u ∧
Continuous (fun x => (1 + u x)⁻¹ - 1 + u x) ∧
K24.Global.LocalSobolevMem r (fun x => (1 + u x)⁻¹ - 1 + u x) ∧
K24.Global.localSobolevENN r (fun x => (1 + u x)⁻¹ - 1 + u x) ≤
(C : ℝ≥0∞) * K24.Global.localSobolevENN r u ^ (2 : ℕ)) ∧
( ∀ (r : ℝ), 2 < r → ∃ δ : ℝ, 0 < δ ∧ ∃ C : ℝ≥0,
∀ u : K24.Global.FourSpace → K24.Global.Mat4,
(∀ i j, Continuous (fun x => u x i j) ∧
K24.Global.LocalSobolevMem r (fun x => u x i j)) →
K24.Global.matrixSobolevENN r u < ENNReal.ofReal δ →
(∀ x, (1 + u x).det ≠ 0) ∧
Continuous (fun x => (1 + u x)⁻¹ - 1) ∧
(∀ i j, Continuous (fun x => ((1 + u x)⁻¹ - 1) i j) ∧
K24.Global.LocalSobolevMem r (fun x => ((1 + u x)⁻¹ - 1) i j)) ∧
K24.Global.matrixSobolevENN r (fun x => (1 + u x)⁻¹ - 1) ≤
2 * K24.Global.matrixSobolevENN r u ∧
Continuous (fun x => (1 + u x)⁻¹ - 1 + u x) ∧
(∀ i j, Continuous (fun x => ((1 + u x)⁻¹ - 1 + u x) i j) ∧
K24.Global.LocalSobolevMem r (fun x => ((1 + u x)⁻¹ - 1 + u x) i j)) ∧
K24.Global.matrixSobolevENN r (fun x => (1 + u x)⁻¹ - 1 + u x) ≤
(C : ℝ≥0∞) * K24.Global.matrixSobolevENN r u ^ (2 : ℕ)) := by sorry