A positive sharp loss has a derived orthogonal spectral gap
ProvedConnesGreen.canonical_positive_threshold_orthogonal_gapconnes-weilfinite-supportfull-tailsharp-threshold
Fix the ORIGINAL positive support window and finite actual-zero selected packet. If its sharp loss is positive, the ORIGINAL sharp shifted margin has a POSITIVE uniform gap on the orthogonal complement of the actual maximizing eigenspace , where :
The spectral gap is derived from original compactness and the sharp bound, not postulated. Actual-zero multiplicities and the complete positive actor remain unchanged.
Preamble
import Definitions.Def_ConnesGreen_positive_endpoint_support import Mathlib.Analysis.InnerProductSpace.Spectrum import Mathlib.Analysis.InnerProductSpace.StarOrder import Mathlib.Analysis.CStarAlgebra.ContinuousFunctionalCalculus.Order import Mathlib.Analysis.Normed.Operator.Compact.FiniteDimension import Mathlib.LinearAlgebra.Eigenspace.ContinuousLinearMap import Theorems.Thm_ConnesGreen_canonicalPositiveCovariance_compact import Theorems.Thm_ConnesGreen_canonical_selected_form_ge_overlap_shift set_option autoImplicit false set_option maxHeartbeats 3000000 set_option synthInstance.maxHeartbeats 200000 set_option backward.isDefEq.respectTransparency false open Complex MeasureTheory ConnesRZ ConnesRZFrontier ContinuousLinearMap Set Filter open scoped BigOperators InnerProductSpace lp ENNReal Classical ComplexOrder Topology noncomputable section open ConnesGreen WeilDefect WeilDefect.ConnesNative ContinuousFunctionalCalculus WeilDefect.MarkerStability
Formal statement
theorem ConnesGreen.canonical_positive_threshold_orthogonal_gap (t : ℝ) (ht : 0 < t)
(S : Finset CriticalZeros) (hμ : 0 < canonicalCertificateThreshold t ht S) :
∃ c : ℝ, 0 < c ∧ ∀ x : Physical t,
x ∈ (canonicalMaximizingSpace t ht S)ᗮ →
c * ‖x‖ ^ 2 ≤ canonicalSharpMargin t ht S x := by sorry
Source
monocap-tech/weil, WeilDefect/Connes/PositiveEndpointSupport.lean, exact declaration ConnesGreen.canonical_positive_threshold_orthogonal_gap, compiling source c9aab301faeba96dcda203002677d6cf2cf396c6