Pointwise reconstruction by the biased Fourier expansion
ProvedNarrowDNF.Fourier.expansioncombinatoricsnarrow-dnf-authors-importprobabilitystoc-2027
Let be finite, , and . For each , let
Then for every input ,
This is exact finite-dimensional reconstruction, with all subsets retained and no limiting or approximation argument. In particular, it holds for arbitrary real-valued input functions. The support subsets are equivalently the Boolean masks indexing the formal basis.
Preamble
import Mathlib.Algebra.BigOperators.GroupWithZero.Finset
import Mathlib.Algebra.BigOperators.Ring.Finset
import Mathlib.Algebra.Order.BigOperators.Group.Finset
import Mathlib.Algebra.Order.BigOperators.Ring.Finset
import Mathlib.Analysis.SpecialFunctions.Sqrt
import Mathlib.Data.Fintype.BigOperators
import Mathlib.Data.Fintype.Pi
import Mathlib.Data.Real.Basic
import Mathlib.Logic.Equiv.Prod
import Mathlib.Tactic.FieldSimp
import Mathlib.Tactic.Linarith
import Mathlib.Tactic.Ring
import Definitions.Def_NarrowDNF_Fourier
import Definitions.Def_NarrowDNF_Influence
import Definitions.Def_NarrowDNF_ProductMeasure
import Theorems.Thm_NarrowDNF_Fourier_kernel
namespace NarrowDNF.FiberShift
end NarrowDNF.FiberShift
namespace NarrowDNF.Fourier
end NarrowDNF.Fourier
namespace NarrowDNF.Influence
end NarrowDNF.Influence
namespace NarrowDNF.ProductMeasure
end NarrowDNF.ProductMeasure
noncomputable section
open scoped BigOperators
open NarrowDNF
open NarrowDNF.Fourier
section Cube
variable {ι : Type*} [Fintype ι] [DecidableEq ι]
Formal statement
theorem NarrowDNF.Fourier.expansion {p : ℝ} (hp0 : 0 < p) (hp1 : p < 1)
(f : (ι → Bool) → ℝ) (x : ι → Bool) :
(∑ s : ι → Bool, coefficient p f s * character p s x) = f x := by sorrySource