Alphabeta Math
TheoremStatement: AI-adaptedProof: AI-adaptedprecheck passjudge pass (gpt-6-luna)
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.

For f:A→R the set of points of A at which f is discontinuous is the intersection with A of an Fσ subset of R, and the set of points at which f is continuous is the intersection with A of a Gδ subset; for A=R the two sets are Fσ and Gδ outright

Statement

Throughout, N⊆Z⊆Q⊆R denotes the canonical copies. Natural numbers, integers and rationals are written directly when used in real arithmetic.

Let A⊆R and let f:A→R. Write

D  :=  { x∈A:f is discontinuous at x },C  :=  A∖D

(Continuity of f:A→R at a point of A and on A: the ε-δ condition, its agreement with lim⁡x→cf(x)=f(c) at a limit point, and continuity at an isolated point, Discontinuity at a domain point: removable, jump and essential cases, and the first- and second-kind grouping). Then:

  1. Pointwise exhaustion. D={ x∈A:ωf(x)>0 } (The oscillation ωf(S)=sup⁡{ ∣f(x)−f(y)∣:x,y∈S } of f on a set and the oscillation ωf(c)=inf⁡δ>0ωf(A∩Nδ(c)) at a point, both taken in the extended reals), and D is the union of the increasing sequence of superlevel sets D  =  ⋃n∈N{ x∈A:ωf(x)≥1/(n+1) } (The canonical natural ι(n)=n⋅1F of a field), whose thresholds are 1,1/2,1/3,….
  2. Descriptive form. There is an Fσ set F⊆R and a Gδ set V⊆R (Fσ and Gδ subsets of R) with D  =  A∩F,C  =  A∩V,V=R∖F, and F may be taken to be ⋃n∈NGn with each Gn a closed subset of R cutting down on A to the n-th set of claim 1 (For every real ε>0 the set { x∈A:ωf(x)≥ε } is the intersection with A of a closed subset of R; in particular it is closed in R when A=R).

In particular, when A=R the discontinuity set D is an Fσ subset of R and the continuity set C is a Gδ subset, and claim 1 reads D=⋃n{ x∈R:ωf(x)≥1/(n+1) }.

Claim 1 is stated separately because it is what later arguments cite downstream. The exhaustion of D by the superlevel sets {ωf≥1/(n+1)} is exactly the form needed when a later proof establishes a property one threshold at a time, so the identity itself is recorded here and not only the descriptive conclusion of claim 2.

The statement is relative on purpose. For a general domain A the sets D and C are subsets of A, and neither is Fσ or Gδ in R in general; what the proof produces are two subsets of R that cut down to them. The absolute form is stated only for A=R, which is the case used later by the realization and exact-continuity-set examples.

Facts & Assumptions

Given: A⊆R and a function f:A→R.

[L2]
[L3]

For every real η>0 there is a natural m≥1 with 1/m<η, where m is identified with its canonical real value; and the natural numbers ≥1 are positive in R and their order is preserved there (For every ε>0 in a complete ordered field there is a natural n≥1 with 1/n<ε, The canonical natural ι(n)=n⋅1F of a field, Canonical naturals are positive and strictly increasing).

[L4]

A subset of R is Fσ when it is the union of a sequence of closed sets and Gδ when it is the intersection of a sequence of open sets; S is Fσ if and only if R∖S is Gδ (Fσ and Gδ subsets of R, Open subset of R (every point has a neighbourhood inside it), closed subset (complement open), and clopen).

Proof

technique · direct
1.1L2L3construct

For each n∈N put εn:=1/(n+1), a positive real since n+1≥1, and let Gn⊆R be closed with {x∈A:ωf(x)≥εn}=A∩Gn.

1.2L1

D={ x∈A:ωf(x)>0 }: a point x∈A is a discontinuity exactly when ωf(x)≠0, and ωf(x)≥0 always, so exactly when ωf(x)>0.

2.1step 1.1step 1.2L3

D⊆⋃n∈N(A∩Gn). Let x∈D, so ωf(x)>0. If ωf(x)≥ε0=1 then x∈A∩G0. Otherwise 0<ωf(x)<1, so ωf(x) is a positive real, and there is a natural m≥1 with 1/m<ωf(x); writing m=n+1 with n∈N gives ωf(x)>εn, hence x∈A∩Gn.

2.2step 1.1step 1.2L3

Conversely ⋃n∈N(A∩Gn)⊆D: if x∈A∩Gn then ωf(x)≥εn>0, so x∈D.

3.1step 1.1step 2.1step 2.2L4

Put F:=⋃n∈NGn, an Fσ subset of R since each Gn is closed and the family is indexed by N. Then A∩F=⋃n(A∩Gn)=D.

3.2step 1.1step 1.2step 2.1step 2.2L3

Claim 1 is proved: D={x∈A:ωf(x)>0} by step 1.2, and D=⋃n∈N{x∈A:ωf(x)≥εn} by steps 2.1 and 2.2, since A∩Gn is by step 1.1 exactly the set {x∈A:ωf(x)≥εn} with εn=1/(n+1). The union is increasing, since n≤m gives (n+1)≤(m+1) and hence εm≤εn.

4.1step 3.1L4

Put V:=R∖F, a Gδ subset of R. Then A∩V=A∖(A∩F)=A∖D=C.

5.1step 3.1step 3.2step 4.1∎

Claim 2 is proved by steps 3.1 and 4.1; and for A=R the two identities read D=F and C=V, so D is Fσ and C is Gδ outright.

Remarks

Depends on

Dependency tree · 2 levels

Sources