Corollary 8.6 — supp νₘ ⊆ [−ρₘ, ρₘ] for random regular graphs
ProvedStrongConv.supp_subset_Icc_rhoStair_of_extends_nuAdjLet and . If is a compactly supported distribution (Definition 4.6) that agrees on real polynomials with the -th coefficient of the expansion of in powers of (nuAdj d m), then . Here if and otherwise. The paper's are the distributions of Theorem 7.1, the unique extensions (Lemma 4.7); the statement is phrased for every extension.
import Mathlib import Definitions.Def_StrongConv open Polynomial
namespace StrongConv
theorem supp_subset_Icc_rhoStair_of_extends_nuAdj (d : ℕ) (hd : 2 ≤ d) (m : ℕ) (hm : m ≤ d)
(μ : smoothFun →ₗ[ℝ] ℝ) (hμ : IsCSDistribution μ)
(hext : ∀ h : ℝ[X], μ (polyFun h) = nuAdj d m h) :
supp μ ⊆ Set.Icc (-rhoStair d m) (rhoStair d m) := by
sorry
end StrongConvRead-back
What the Lean code literally says, in plain math · claude-opus-5-5
What the statement asserts
Let and be natural numbers with and (so is allowed). Let be a real-linear functional on , the real vector space of all infinitely differentiable functions (no support or growth condition). Assume:
- is a finite-order distribution controlled on a compact interval: there are reals , and a natural number with
- agrees with the -th asymptotic coefficient on polynomials: for every real polynomial , viewed as the smooth function ,
with as defined below (it depends on ).
Then
a closed interval, where
All arithmetic here is in the real numbers ( when , no truncation), and is the nonnegative square root. Since , ; in the second case , so the denominator is positive; and in both cases. For the first case applies, giving .
The objects, in full
Support. Say that a set carries if for every that vanishes identically on some open set containing . is the intersection of all closed sets that carry (the whole line always carries , so the family is nonempty).
The random operator . For let , the orthogonal complement of the all-ones vector in (dimension for ; ). For a permutation of , is the permutation operator with , and its restriction to . For ,
with the adjoint within . is self-adjoint, and for a polynomial , is the usual polynomial in (formally defined by the continuous functional calculus applied to , which agrees with it).
The sequence . For a real polynomial and ,
the average over independent uniform of the trace on divided by (not ); division by gives , so .
The coefficients . Defined recursively for by
limits over natural numbers ; in particular and . If a limit does not exist, the corresponding value is an unspecified real number (a fixed but arbitrary choice made by the formal system), not and not "undefined", and later coefficients are computed from that value.
Edge cases and caveats
- is included: hypothesis 2 then says , and the interval is .
- is the largest allowed value; is excluded by hypothesis.
- Hypothesis 2 may be unsatisfiable. is linear, so hypothesis 2 forces to be linear on polynomials and to obey the bound of hypothesis 1 there. If some limit in the recursion fails to exist, the values are unspecified, and whether any meets hypothesis 2 is not determined by the statement; when no such exists the statement holds vacuously.
- is fixed only on polynomials. On other smooth functions it is constrained only by linearity and hypothesis 1.
- The constant of hypothesis 1 does not appear in the conclusion.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.