Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-generatedverified 2026-09-23 (gpt-6-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.

Differentiation under the integral sign

Statement

Let I⊆R be an open interval and let f:X×I→C be such that:

  1. for every t∈I, the function x↦f(x,t) is integrable;
  2. for almost every x, the map t↦f(x,t) is differentiable on I;
  3. for every t∈I, the function x↦∂f∂t(x,t), extended by zero where the derivative is undefined, is measurable;
  4. there are a measurable null set N and a nonnegative measurable function g with ∫g dμ<+∞ and ∣∂f∂t(x,t)∣≤g(x) for every t∈I and every x∈X∖N.

Then F(t):=∫f(x,t) dμ(x) is differentiable on I, and F′(t)=∫∂f∂t(x,t) dμ(x), with the same zero extension in the last integral.

Facts & Assumptions

Given: An open interval I, a function f satisfying the first three displayed hypotheses, and a measurable null set N together with a nonnegative measurable majorant g satisfying hypothesis 4. Choose a measurable null set E outside which hypothesis 2 holds, and put N∗=N∪E.

[L1]

Dominated convergence applies to integrable complex-valued functions under a single L1 majorant (Dominated convergence).

[L3]

The complex integral is linear on L1 (The Lebesgue integral is linear on L1(μ)).

[L4]

The rationals are dense in the reals (The rationals embed densely in the reals).

Proof

technique · direct
1.1givenconstruct

Fix t0∈I and let hn→0 with hn≠0 and t0+hn∈I. Define qn(x):=f(x,t0+hn)−f(x,t0)hn. For every x∈X∖N∗, differentiability in t gives qn(x)→∂tf(x,t0). Give the derivative value zero on N∗; this measurable modification differs from the stated zero extension only on a null set, so it has the same integral.

2.1step 1.1L1L2

For each n, hypothesis 1 makes x↦f(x,t0+hn) and x↦f(x,t0) integrable and therefore measurable, so qn is measurable. Fix x∈X∖N∗ and n. If qn(x)=0, then ∣qn(x)∣≤g(x) is immediate. Otherwise put α:=qn(x)‾/∣qn(x)∣, so ∣α∣=1 and ∣qn(x)∣=Re⁡ ⁣(α f(x,t0+hn)−f(x,t0)hn). Apply [L2] to the real-valued function τ↦Re⁡(αf(x,τ)) on the segment joining t0 to t0+hn. For some interior point ξ of that segment, ∣qn(x)∣=Re⁡(α ∂tf(x,ξ))≤∣∂tf(x,ξ)∣≤g(x). Hypothesis 3 and the zero extension make the limit measurable. Therefore [L1] applies to (qn).

3.1step 2.1L1L3

By [L1], lim⁡n∫qn(x) dμ(x)=L:=∫∂tf(x,t0) dμ(x). Linearity of the integral gives ∫qn(x) dμ(x)=F(t0+hn)−F(t0)hn. This holds for every supplied sequence of admissible nonzero increments.

4.1L2L4step 2.1step 3.1∎

The same mean-value estimate as in step 2.1, with any s,t∈I in place of t0,t0+hn, gives ∣f(x,t)−f(x,s)∣≤∣t−s∣g(x) outside N∗. Integrating yields ∣F(t)−F(s)∣≤∣t−s∣∫g dμ, so F is continuous. Consequently Q(h)=(F(t0+h)−F(t0))/h is continuous on the punctured interval of admissible increments. If Q(h) failed to tend to L as h→0, there would be an ε>0 and, for every n≥1, a nonzero admissible h with ∣h∣<1/n and ∣Q(h)−L∣≥ε. Continuity of Q and [L4] give a rational admissible r with ∣r∣<1/n and ∣Q(r)−L∣>ε/2. Fix an enumeration of Q and take the first such rational rn for each n; this is a definable selection from a countable set and uses no Countable Choice. Then rn→0, contradicting step 3.1. Thus Q(h)→L, which is precisely F′(t0)=L. Since t0 was arbitrary, the theorem follows.

Depends on

Used by

…and 1 more result.

Dependency tree · two levels

25 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