Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-generatedPipeline-generatedjudge pass (gpt-5.6-terra)audited 2026-09-12
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.

Probability algebras, arbitrary joins and the countable chain condition

Statement

Assume AC. Let (X,Σ,μ) be a probability space, so μ(X)=1. Identify A,B∈Σ when μ(A△B)=0, and write B for the set of equivalence classes. Set operations induce Boolean operations on B, and m([A])=μ(A) is a well-defined strictly positive probability on it. The order is [A]≤[B] exactly when μ(A∖B)=0.

The Boolean algebra B is complete. Every subset D⊆B has a countable subset D0⊆D with ⋁D=⋁D0, allowing D0=∅. Every antichain of nonzero elements is countable. For arbitrary D⊆B and c∈B,

c∧⋁D=⋁{c∧d:d∈D}.

Countable joins are represented by countable unions of measurable representatives. No assertion that an arbitrary union of representatives is measurable or represents its Boolean join is made.

Facts & Assumptions

Given: A probability space and AC. All families below are sets.

[F1]

Measures are countably additive on disjoint measurable sequences. (Measures on sigma-algebras)

[F2]

Countable unions of measurable null sets are null, by countable subadditivity. (Finite and countable subadditivity of measures)

[F3]

The measure of an increasing measurable union is the supremum of the measures. (Continuity from below for measures)

[F4]

Boolean algebras have the bounded distributive lattice laws, with order given by meet. (Boolean algebras and their order)

[F5]

Completeness means existence of every set supremum, including the empty supremum zero. (Completeness, regular opens, and order continuity)

[F6]

AC selects measurable representatives, countably many finite approximants to a supremum, and enumerations of countably many finite antichain pieces. (The Axiom of Choice)

Proof

1.1F1F2F4

Symmetric difference is symmetric, A△A=∅, and A△C⊆(A△B)∪(B△C). F1 and F2 therefore give an equivalence relation. Its classes form a set quotient of Σ. Replacing either input of a union or intersection by an equivalent set changes the result only inside the union of the input symmetric differences; complement preserves symmetric difference. Hence the operations are well-defined and inherit all F4 identities from set operations. Splitting sets into disjoint differences shows that equivalent sets have equal measure, so m is well-defined. It vanishes exactly on the zero class, and m(1)=1, making the algebra nontrivial. The equation [A]∧[B]=[A] is equivalent to μ(A∖B)=0.

2.1F1F2F6step 1.1

For a sequence (an) choose representatives An by F6 and put a=[⋃nAn]. This bounds every an. If b=[B] is another upper bound, each An∖B is null; F2 makes their union null, so a≤b. Thus a is the supremum. Another sequence of representatives gives the same class, again by F2. For a disjoint sequence of Boolean elements, remove from An all earlier Aj to obtain literally disjoint representatives: the removed part is a finite union of null intersections. F1 then proves m(⋁nan)=∑nm(an). An empty sequence has supremum zero.

3.1F3F6F7step 2.1

Given D⊆B, let s be the supremum supplied by F7 in [0,1] of m(⋁F) over finite F⊆D, including F=∅. Choose finite Fn⊆D with m(⋁Fn)>s−2−n; when s=0 all Fn may be empty. Put D0=⋃nFn, which is countable by F6, and let c=⋁D0 using step 2.1. The finite joins over F0∪⋯∪Fn increase to c. Measurable representatives can be chosen increasing by taking successive finite unions, so F3 gives m(c)=s.

3.2F6step 1.1step 2.1

Let A be an antichain of nonzero elements. For each positive integer n put An={a∈A:m(a)≥1/n}. Any n+1 distinct members of An would have disjoint join of measure at least (n+1)/n>1, contrary to step 2.1. Thus An has at most n elements. Strict positivity gives A=⋃n≥1An, and F6 makes this union countable. The empty antichain and singleton antichains satisfy the same bound.

4.1F1F3F5step 1.1step 2.1step 3.1

For any d∈D, the finite joins in step 3.1 with d adjoined still have measure at most s. F3 gives m(c∨d)≤s=m(c). Disjoint additivity then gives m(d∧¬c)=m(c∨d)−m(c)=0, hence d≤c by strict positivity. Any upper bound of D bounds D0 and thus bounds c by step 2.1. Therefore c=⋁D, proving completeness and the countable-subfamily assertion. If D is empty the construction gives c=0; if it is a singleton its supremum is that element.

5.1F4step 4.1∎

Put u=⋁D and v=⋁{c∧d:d∈D}. Each joined term lies below c∧u, so v≤c∧u. Conversely c∧d≤v implies d=(c∧d)∨(¬c∧d)≤v∨¬c. Hence u≤v∨¬c, and finite distributivity gives c∧u≤c∧(v∨¬c)≤v. This proves the identity, including c=0, c=1 and D=∅. Only the established Boolean supremum is used, not the possibly nonmeasurable union of an arbitrary family of representatives.

Depends on

Used by

Dependency tree · two levels

23 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