Alphabeta Math
ExampleConstruction: AI-generatedVerification: AI-generatedPipeline-generatedprecheck passjudge pass (gpt-6-luna)verified 2026-10-22 (gpt-6.1-sol)
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.

Conditional stability on a power-source class

Example

Assume Countable Choice. Let K:X→Y be a compact operator of infinite rank between real or complex Hilbert spaces over one scalar field. Let (sj,vj,uj)j≥1 be a positive singular system with sj>0, sj→0, ∥vj∥X=∥uj∥Y=1, and Kvj=sjuj. For ρ>0 let Mρ=X1,ρ(K)={∣K∣w:w∈X, ∥w∥X≤ρ} and define εK(Mρ,δ)=sup⁡{∥x∥X:x∈Mρ, ∥Kx∥Y≤δ}. Then εK(Mρ,δ)≤ρδ for every δ≥0, with equality at δj=ρsj2 for each j, witnessed by xj=ρsjvj. Although K† is discontinuous on its full domain, this modulus tends to zero on the restricted source class Mρ; this conditional stability makes no claim of universal noise-level-free convergence.

Facts & Assumptions

Given: Countable Choice, a compact infinite-rank operator K:X→Y between real or complex Hilbert spaces over one scalar field, a positive singular system (sj,vj,uj)j≥1, ρ>0, the class Mρ=X1,ρ(K), and δ≥0.

[A1]

The Axiom of Countable Choice (ACω): Countable Choice is assumed by the compact singular-value decomposition and Hilbert-space pseudoinverse definitions.

[F1]

Compact linear inverse model between Hilbert spaces: The operator is compact and linear between real or complex Hilbert parameter/data spaces over one field.

[F2]

Singular value decomposition for compact operators: The positive singular directions are orthonormal, with vj∈N(K)⊥, ∣K∣vj=sjvj, and Kvj=sjuj for each positive singular value sj>0.

[F3]

Hilbert-space Moore–Penrose inverse: D(K†)=R(K)⊕R(K)⊥, and K† returns the minimum-norm preimage of the range component.

[F4]

Power source ball: Mρ=X1,ρ(K)={∣K∣w:w∈X, ∥w∥X≤ρ}.

[F5]

Centered conditional modulus of continuity: εK(Mρ,δ) is the supremum of ∥x∥X over x∈Mρ with ∥Kx∥Y≤δ.

[F6]

Conditional modulus on a power source ball: For ν=1, εK(X1,ρ(K),δ)≤δ1/2ρ1/2; equality holds at δj=ρsj2 with witness xj=ρsjvj.

[F7]

Bakushinskii veto for discontinuous pseudoinverses: For discontinuous K†, universal worst-case convergence over every exact datum and every admissible noise ball would imply continuity of K†.

Verification

1.1givenA1F1F2F3F6algebra

By [F6] with ν=1, the equality radii δj=ρsj2 form a null sequence; since ρ>0, sj→0. Put yj:=Kvj=sjuj. Then ∥yj∥Y=sj→0 by [F2]. Since vj∈N(K)⊥, [F3] gives K†yj=vj, whose norm is 1. Thus K† is discontinuous at 0 on its domain.

2.1givenstep 1.1A1F1F2F4F5F6F7constructalgebra∎

By [F4]–[F6] with ν=1, εK(Mρ,δ)≤δ1/2ρ1/2, and for each j, xj=∣K∣(ρvj)=ρsjvj lies in Mρ, has ∥Kxj∥Y=ρsj2=δj, and has ∥xj∥X=ρsj=ρδj. This attains the bound. Since δj→0 by [F6], the upper bound tends to zero as δ↓0. The estimate is restricted to Mρ and does not assert the universal convergence condition in [F7].

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

47 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