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.

Measurable dense selections for fields of nonempty compact sets

Statement

Assume the Axiom of Choice (The Axiom of Choice). Let (X,A,μ) be a standard Borel space with a sigma-finite measure (Standard Borel spaces, Measure spaces, Finite, sigma-finite, and semifinite measures). Let (K,d) be a nonempty compact metric space (Open cover, subcover, and compact topological space; a compact subset is a subspace that is compact in its own right, Topology on a set, open and closed sets, clopen sets, the closed-set axiomatisation, and the coarser/finer comparison, Metric space: d(x,y)=0 iff x=y, symmetry, and the triangle inequality; pseudometric and ultrametric, The metric topology: a set is open when every one of its points has a ball around it inside the set; closed means open complement) with a fixed dense sequence (km)m∈N, and give K its Borel sigma-algebra (The Borel sigma-algebra of a topological space). Suppose g:X×K→[0,∞) has measurable sections x↦g(x,k) for each fixed k, continuous sections k↦g(x,k) for each fixed x, and nonempty zero sets Cx:={k∈K:g(x,k)=0} for every x. Then: (1) there is a sequence of measurable maps sj:X→K such that sj(x)∈Cx for all x,j and (sj(x))j∈N is dense in Cx for every x; (2) for every k∈K, the function x↦d(k,Cx):=inf⁡c∈Cxd(k,c) is measurable; and (3) for every open U⊆K, the hit set {x:Cx∩U≠∅} is measurable.

Facts & Assumptions

Given: The measurable space (X,A) is standard Borel, μ is sigma-finite, (K,d) is compact with its metric topology, the dense sequence (km) is fixed, and g has the stated measurable and continuous sections with nonempty zero sets.

[F1]

A measurable space has a sigma-algebra of measurable sets; it is closed under countable unions and intersections. A standard Borel space is in particular a measurable space. The Borel sigma-algebra of K is generated by its open sets and is minimal among sigma-algebras containing them (Standard Borel spaces, Measure spaces, Measures on sigma-algebras, Finite, sigma-finite, and semifinite measures, Measurable spaces and measurable sets, A measurable function between measurable spaces, The Borel sigma-algebra of a topological space, Sigma-algebras, Nonempty intersections of sigma-algebras are sigma-algebras, so the generated sigma-algebra exists and is minimal).

[F3]

Metric distances are nonnegative; every nonempty subset of N has a least element; and for each ρ>0 there is n≥1 with 2/n<ρ, by applying the reciprocal bound to ρ/2 (Order on the reals, Complete ordered field (least-upper-bound property), Maximum and minimum of a set, Nonnegativity of a metric is a consequence of the other axioms, not an axiom, For every ε>0 in a complete ordered field there is a natural n≥1 with 1/n<ε, The well-ordering principle).

[F4]

Measurable maps compose with Borel maps; pointwise sums, products, absolute values, maxima and minima of real measurable functions are measurable; and real-valued measurability follows from measurability of all strict sublevel sets (A measurable function between measurable spaces, Composition with a Borel measurable outer map preserves measurability, Arithmetic and lattice operations preserve measurability whenever they are defined, Threshold characterisations of real-valued and extended-real-valued measurability).

[F5]

AC is the explicit hypothesis in the Statement (The Axiom of Choice). The proof uses no choice: (km) is given, recursive indices are least natural numbers, and every limit selector is unique. The sigma-finite measure is not used.

Proof

technique · direct
1.1F1F2F3

First prove a closed-target hit claim for any field h:X×K→[0,∞) with measurable x-sections, continuous K-sections, and nonempty zero fibers Zx:={k:h(x,k)=0}. For a closed F⊆K and n≥1, let MF,n:={m≥0:∃y∈F, d(km,y)<1/n}. Then {x:Zx∩F≠∅}=⋂n≥1 ⋃m≥0: m∈MF,n{x:h(x,km)<1/n}, where each union is read as an N-indexed union with empty terms for m∉MF,n. If F=∅, both sides are empty. If z∈Zx∩F, continuity at z and density of (km) provide, for each n, an m with d(km,z)<1/n and h(x,km)<1/n, so the right side holds. Conversely, suppose the right side holds and put Ar:={km:m∈MF,r, h(x,km)<1/r}; each Ar is nonempty. The closed sets Ln:=⋃r≥nAr‾ are nonempty and nested. They have the finite intersection property, so [F2] gives z∈⋂nLn. For any ρ>0, choose n with 2/n<ρ by [F3]; since z∈Ln, some km∈Ar for an r≥n satisfies d(km,z)<1/n, and m∈MF,r gives a y∈F with d(km,y)<1/r≤1/n. Thus d(z,y)<2/n<ρ, so z∈F. Given ϵ>0, continuity at z gives δ>0 such that d(w,z)<δ implies ∣h(x,w)−h(x,z)∣<ϵ/2. Choose n with 1/n<min⁡(δ,ϵ/2) by [F3]. Some km∈Ar for r≥n then satisfies d(km,z)<1/n and h(x,km)<1/r≤1/n<ϵ/2, so 0≤h(x,z)<ϵ. Since this holds for every ϵ>0, h(x,z)=0 and z∈Zx∩F. The displayed set is measurable by [F1] and the measurable-section hypothesis.

2.1F1F3F4step 1.1

Step 1.1 shows that every closed-target hit set is measurable. If U⊆K is open, it is the union of the countable family of closed balls Bˉ(km,1/n) that are contained in U: for y∈U, choose ρ>0 with B(y,ρ)⊆U, choose n with 2/n<ρ, then choose m with d(km,y)<1/(2n). This gives y∈B(km,1/n) and Bˉ(km,1/n)⊆B(y,ρ)⊆U. Thus {x:Zx∩U≠∅}=⋃m≥0 ⋃n≥1: Bˉ(km,1/n)⊆U{x:Zx∩Bˉ(km,1/n)≠∅}, an iterated countable union, so open-target hit sets are measurable by [F1] and step 1.1. For fixed k∈K and q>0, d(k,Zx)<q exactly when Zx meets B(k,q), by the definition of infimum; for q≤0 the strict sublevel set is empty by nonnegativity. The infimum exists in R because these distances form a nonempty set bounded below by 0 and R is complete; it is finite because any one point of Zx gives a finite upper bound. Hence every strict sublevel set of x↦d(k,Zx) is measurable, and [F4] proves that this distance function is measurable.

3.1F1F2F3F4step 2.1

Apply steps 1.1–2.1 to a field h as above, and put G1=h, Zx1=Zx. Since each h(x,⋅) is continuous, each initial zero fiber is closed. For n≥2, define mn(x) to be the least m≥0 such that d(km,Zxn−1)<1/n, and set Gn(x,k):=Gn−1(x,k)+max⁡{d(k,kmn(x))−1/n,0},Zxn:={k:Gn(x,k)=0}=Zxn−1∩Bˉ(kmn(x),1/n). This zero-set identity uses that both summands are nonnegative. Such an m exists by density of (km) and nonemptiness of Zxn−1. For M≥0, {x:mn(x)≤M}=⋃m=0M{x:Zxn−1∩B(km,1/n)≠∅}, which is measurable by step 2.1 applied to Gn−1. Differences give measurable singleton fibers of mn; for any S⊆N, mn−1(S) is the countable union of those fibers, taking the empty set for indices outside S. Equip N with its power-set sigma-algebra; then mn is measurable, and so is x↦kmn(x) by composition, since every map from this discrete measurable space into K is measurable. For fixed k, the scalar function m↦max⁡{d(k,km)−1/n,0} is measurable on the discrete space, so [F4] makes the added term and then Gn(⋅,k) measurable. For fixed x, k↦d(k,kmn(x)) is continuous by the triangle inequality, and t↦max⁡{t−1/n,0} is continuous because its two affine formulas agree at 1/n; thus Gn(x,⋅) is continuous. Its zero fiber is nonempty because d(kmn(x),Zxn−1)<1/n means that Zxn−1 meets the open ball, and it is closed as the intersection of a closed zero fiber with a closed ball. Its diameter is at most 2/n.

4.1F1F2F3step 2.1step 3.1

For each x, the nested nonempty closed sets Zxn have the finite intersection property, so [F2] gives a point σ(x)∈⋂n≥1Zxn. The diameter bound makes it unique: for every n≥2, any two points in the intersection are at distance at most 2/n, and [F3] makes these bounds arbitrarily small. Since σ(x)∈Zxn⊆Bˉ(kmn(x),1/n), the zero-indexed sequence aq(x):=kmq+2(x) converges to σ(x); also σ(x)∈Zx1=Zx. For open U⊆K, σ−1(U)=⋃m≥0 ⋃n≥1: Bˉ(km,1/n)⊆U ⋃N≥2 ⋂q≥N{x:kmq(x)∈Bˉ(km,1/n)}. These are successive countable unions and intersections over natural indices. Eventual membership in one of these closed balls forces the limit into U. Conversely, if σ(x)∈U, step 2.1 supplies a closed ball contained in U with σ(x) in its open ball, and convergence makes the sequence eventually lie in that closed ball. Each set on the right is measurable because x↦kmq(x) is measurable and the ball is Borel; all unions and intersections are countable. Since open sets generate the Borel sigma-algebra, [F1] proves that σ:X→K is measurable. This constructs an everywhere selection for every admissible field h.

5.1F1F2F3F4step 1.1step 4.1

For m∈N and j≥1, let Xm,j:={x:Cx∩Bˉ(km,1/j)≠∅}, which is measurable by step 1.1, and define hm,j(x,k):=g(x,k)+1Xm,j(x)max⁡{d(k,km)−1/j,0}. Its fixed-k sections are measurable by [F4]; its fixed-x sections are continuous. Its zero fiber is Cx∩Bˉ(km,1/j) on Xm,j and Cx off that set, so it is nonempty for every x. Applying step 4.1 to this field gives a measurable tm,j:X→K with tm,j(x)∈Cx for every x, and tm,j(x)∈Bˉ(km,1/j) when x∈Xm,j. Given y∈Cx and ϵ>0, choose j with 2/j<ϵ and m with d(km,y)<1/(2j). Then x∈Xm,j and d(tm,j(x),y)≤d(tm,j(x),km)+d(km,y)≤1/j+d(km,y)<3/(2j)<2/j<ϵ. Therefore the family (tm,j)m∈N, j≥1 is dense in each fiber. Enumerating pairs (m,j)∈N×N>0 by increasing sum, and within each finite diagonal by increasing first coordinate, gives a sequence (sq)q∈N of measurable selections dense in every Cx.

6.1F5step 2.1step 4.1step 5.1∎

Steps 4.1–5.1 prove assertion (1), and step 2.1 proves (2) and (3). The proof spends no form of Choice: the given dense sequence is fixed input, each recursive index is the least admissible natural number, and every selected limit point is unique. AC remains an explicit but unused hypothesis, and the sigma-finite measure is likewise unused.

Depends on

Used by

Dependency tree · two levels

62 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