Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passaudited 2026-10-08
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Monomials form complete orthogonal systems of the Bergman spaces of the disc, the ball and the polydisc

Statement

Assume ACω (The Axiom of Countable Choice (ACω)) and let m≥1. For k≥0 and α∈Nm, the monomials in each of A2(D), A2(Bm) and A2(Dm) are pairwise orthogonal. Their squared norms are, respectively, ∥zk∥A2(D)2=πk+1,∥zα∥A2(Bm)2=πmα!(m+∣α∣)!,∥zα∥A2(Dm)2=πm∏j<m(αj+1). Thus the normalized monomials form complete orthonormal systems in all three Bergman spaces. Equivalently, for each of these domains Ω, a function f∈A2(Ω) orthogonal to every monomial is identically zero.

Facts & Assumptions

[A1]

The only choice principle is ACω: it enters through the Bergman Hilbert structure, monomial norms, real-linear change of variables for rotations, Borel-to-Lebesgue measurability, and the Hilbert-space completeness criterion; no full Axiom of Choice is used (The Axiom of Countable Choice (ACω)).

[F1]

The Bergman definition identifies A2 with holomorphic L2 classes, gives the first-variable-linear integral pairing, and provides their unique holomorphic representatives (The Bergman space A2(Ω) and the Bergman kernel).

[F2]

The monomial square norms on the disc, ball and polydisc are the formulas in the statement; they are positive and finite (Weighted monomial integrals and monomial norms for the disc, ball and polydisc).

[F3]

Each coordinate projection is holomorphic since its increment at a in direction h is hj; finite products of holomorphic functions are holomorphic and holomorphic functions are continuous (Holomorphic functions on an open subset of Cm, Sums, products and nonvanishing quotients of holomorphic functions are holomorphic, A holomorphic function of several variables is continuous and separately holomorphic).

[F4]

On every polydisc centered at 0 whose closure lies in the domain, the Taylor series of a holomorphic function converges absolutely and uniformly on each smaller closed polydisc (A continuous separately holomorphic function is the sum of an absolutely convergent power series with Cauchy-integral coefficients on every smaller polydisc).

[F5]

The Taylor coefficients at 0 are independent of which such centered polydisc is used (The coefficients of a convergent multi-indexed power series are its derivative coefficients, hence unique).

[F7]

Multiplication of one coordinate by η=a+ib acts on its real coordinate pair by (a−bba), whose determinant is ∣η∣2; all other real coordinates are fixed, so the full determinant is also ∣η∣2. If ∣η∣=1, the rotation preserves each of the three domains and its dilates. Applying the linear change-of-variables theorem to the inverse rotation shows that every restricted map is measurable and measure preserving (Complex m-space and its real coordinate dictionary, Real and imaginary parts, complex conjugation, and modulus, Measure-preserving transformations and systems, A linear map T of Rn sends Lebesgue measurable sets to Lebesgue measurable sets, with λn(T[E])=∣det⁡T∣ λn(E) when T is invertible and T[E] Lebesgue null when it is not).

[F8]

Integrals of integrable complex functions are invariant under a measure-preserving map (Integral invariance under measure-preserving maps).

[F10]

A pointwise almost-everywhere limit dominated by one integrable nonnegative function has convergent complex integrals (Dominated convergence).

[F11]

For u,v∈L2(Ω), the product uv‾ is integrable and ∫∣uv‾∣≤∥u∥2∥v∥2 (Cauchy–Schwarz: ∣⟨x,y⟩∣≤∥x∥ ∥y∥, with equality exactly for dependent pairs).

[F13]

The nonnegative integral is monotone; in particular λ(tΩ)≤λ(Ω)<∞ (Monotonicity and nonnegative homogeneity of the nonnegative integral, Weighted monomial integrals and monomial norms for the disc, ball and polydisc).

[F14]

An orthonormal family in a Hilbert space is complete exactly when its orthogonal complement is {0} (Orthonormal families, complete orthonormal systems and Hilbert bases, Parseval equivalences for an orthonormal family).

[F15]

The integral of a finite sum of integrable complex functions is the corresponding finite sum of their integrals (The Lebesgue integral is linear on L1(μ)).

[F16]

If u,v are continuous complex functions, then uv‾ is continuous: near any point a, ∣u(x)v(x)‾−u(a)v(a)‾∣≤∣u(x)∣ ∣v(x)−v(a)∣+∣v(a)∣ ∣u(x)−u(a)∣, and continuity of u bounds it locally; conjugation preserves modulus (Conjugation is an involutive real-field automorphism, zz‾=∣z∣2, and modulus is definite, multiplicative, and subadditive).

[F17]

Under ACω, L2(Ω) is Hilbert and the preceding theorem makes A2(Ω) a closed subspace, hence a Hilbert space (The Bergman space A2(Ω) and the Bergman kernel, A2(Ω) is closed, and the Bergman kernel is the sum over any complete orthonormal system).

Proof

technique · direct, using monomial moments, coordinate rotations, and Taylor coefficients

Given: ACω, one of Ω=D, Bm or Dm (with m=1 in the disc case), and f∈A2(Ω) when testing completeness.

1.1A1F1F2F3given

The preceding monomial-integral lemma gives the three squared-norm formulas in the statement. Every value is positive and finite, so each monomial belongs to the corresponding Bergman space and can be normalized.

1.2A1F1F2F6F7F8F11given

Let α≠β and choose j<m with d0:=βj−αj≠0. Set d=∣d0∣ and η=eiπ/d. The coordinate rotation T(z)j=ηzj, fixing the other coordinates, maps Ω and every tΩ onto themselves and preserves Lebesgue measure by [F7]. For g(z)=zβzα‾, one has g(Tz)=ηβjη‾αjg(z)=ηd0g(z)=−g(z): if d0>0 the factor is ηd=−1, and if d0<0 it is η−d=ηd‾=−1. Invariance [F8] gives I:=∫Ωg=∫Ωg∘T=−I, so I=0 in C. Integrability follows from [F2] and [F11]; therefore ⟨zβ,zα⟩=0. Hence distinct monomials are orthogonal, and the normalized family is orthonormal.

1.3F3F4F5F9given

Fix 0<t<1 and set Kt=tΩ‾. By [F9], Kt is compact and contained in Ω. For each a∈Kt, choose a positive polyradius ρ with Δ‾ρ(0)⊂Ω and ∣aj∣<ρj for all j<m: if Ω=D, take ∣a∣<ρ<1; if Ω=Bm, take ρj>∣aj∣ with ∑j<mρj2<1; and if Ω=Dm, take ∣aj∣<ρj<1. These choices exist since a∈tΩ‾ and t<1. Choose s<ρ and 0<θ<1 with ∣aj∣<θsj for every j<m, possible because there are finitely many strict coordinate inequalities. Holomorphy gives the continuity and separate holomorphy required by [F4]. Define the box partial sums SN(z)=∑α∈{0,…,N}mcαzα. By [F4], these sums converge uniformly on each closed polydisc strictly inside Δρ(0), including Δ‾θs(0). If two admissible radii are used, restrict both expansions to a smaller common centered polydisc and apply [F5]; hence their coefficient families agree. The family of all such open polydiscs Δθs(0) covers Kt, without selecting one for each point. By A subset of a metric space is open in the subspace metric exactly when it is the trace of an open set of the ambient space, and it is compact as a metric space in its own right exactly when every family of open subsets of the ambient space covering it, indexed or not, has finitely many members covering it, this ambient open cover has a finite subcover; a common cutoff for its finitely many uniform convergences shows that these same box partial sums converge uniformly to f on Kt.

2.1A1F1F2F3F8F9F10F11F12F13F15F16step 1.2step 1.3

Fix 0<t<1 and a multi-index γ. The box partial sums SN from step 1.3 converge uniformly on Kt, so they are uniformly bounded there by a finite Mt. Since ∣zγ∣≤1 on Ω and λ(tΩ)<∞ by [F13], the measurable functions 1tΩSNzγ‾ are dominated by the integrable function Mt1tΩ; measurability follows from [F3], [F12] and [F16], and dominated convergence [F10] permits passing their integrals to the limit. For every N≥max⁡jγj, finite-sum linearity [F15] and the rotation argument of step 1.2 give ∫tΩSN(z)zγ‾ dλ2m(z)=cγ∫tΩ∣zγ∣2 dλ2m(z), since every off-diagonal monomial moment vanishes on tΩ by that same rotation argument. Taking the limit yields ∫tΩf(z)zγ‾ dλ2m(z)=cγ∫tΩ∣zγ∣2 dλ2m(z).

3.1A1F1F2F3F10F11F12F16step 2.1given

Suppose f is orthogonal to every monomial. Let tn=1−1/(n+2), so tn↑1 and tnΩ increases to Ω. For each γ, the functions 1tnΩfzγ‾ converge pointwise to fzγ‾ and are dominated by ∣fzγ‾∣, which is integrable by [F11]; measurability follows from [F3], [F12] and [F16]. Also 1tnΩ∣zγ∣2 converges to ∣zγ∣2 and is dominated by it, which is integrable by [F2]; its measurability follows by the same facts. Applying [F10] to both sequences and using step 2.1 gives 0=⟨f,zγ⟩=cγ∥zγ∥22. The norm in [F2] is positive, so cγ=0 for every γ.

4.1F1step 1.3step 3.1given

Step 1.3 supplies the Taylor expansion on each Kt, and the dilates tnΩ exhaust Ω. Since all coefficients vanish by step 3.1, this expansion gives f=0 at every point of Ω. Thus the only element of A2(Ω) orthogonal to every monomial is 0.

5.1A1F1F11F14F17step 1.2step 4.1∎

Step 1.2 makes the normalized monomials an orthonormal family, and step 4.1 makes its orthogonal complement zero. By [F17] the Bergman space is Hilbert, so the zero-complement-to-completeness direction of [F14] shows that this family is complete. Conversely, if the family is complete, every vector orthogonal to its members is orthogonal to their dense linear span and hence, by Cauchy–Schwarz [F11], to itself; it must then be zero. Thus both directions of the stated equivalence hold, and the norm formulas and completeness establish all three asserted complete orthonormal systems.

Depends on

Used by

Dependency tree · two levels

223 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources