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.
Finite partition energy yields Hilbert-valued AC2 paths
Statement
Assume AC. Let be a separable real Hilbert space, , and be continuous. For a finite partition with , set If , then there is a strongly measurable such that for every , is norm-differentiable almost everywhere with derivative , and
Statement references
Facts & Assumptions
Given: AC, , and the finite partition bound .
The interval space with its Bochner inner product is a separable Hilbert space, and interval step functions with values in a countable dense subset of are dense in . Bochner L2 on a finite interval is a separable Hilbert space
A bounded real linear functional on a real Hilbert space has a unique Riesz representative, with equal norm. Riesz representation for Hilbert spaces
Strongly measurable Bochner-integrable functions admit integrable simple approximations, and Bochner integrability is equivalent to integrability of the norm. Bochner-integrable function Bochner integrability criterion
Bochner integrals commute with bounded linear functionals, and Cauchy–Schwarz bounds the scalar pairings. Bounded linear maps commute with Bochner integration Cauchy–Schwarz: , with equality exactly for dependent pairs
AC implies Countable and Dependent Choice, as needed for countable approximations and null-set unions. The Axiom of Choice AC implies DC implies countable choice The Axiom of Countable Choice () The axiom of dependent choice: a relation in which every element is related to something admits an -indexed chain
Scalar locally integrable functions equal the limits of their centered interval averages almost everywhere. Lebesgue differentiation theorem on
Markov's inequality and countable subadditivity show that sets with summable measures have a null limsup. Chebyshev-Markov inequality for the integral Finite and countable subadditivity of measures
Proof
Given: AC, a continuous path , and .
Let be the vector space of all finite -valued interval step functions. It contains the dense family in [F1], hence is dense in . If on a partition , define . Refining leaves this sum unchanged, so is well-defined and linear on . Finite-sum Cauchy–Schwarz gives . By [F1], extends uniquely to a bounded linear functional on , and [F2] represents it by .
Fix and . The function belongs to , so its defining value is . By the Riesz representation in step 1.1 and commutation of bounded linear functionals with Bochner integrals, this equals . The identity holds for every ; take equal to the difference of and to see that its squared norm is zero. Thus in .
By [F3], choose simple -valued functions with and put . Markov's inequality gives ; [F7] shows that outside a null set eventually. Remove also the countable union of differentiation exceptional sets for all scalar functions and the finitely many indicators defining each , using [F5] and [F6]. At each remaining interior , the average of over either one-sided interval with endpoint has limsup at most , since that interval lies in the centered interval of twice its length. The average of tends to zero: has finitely many values and each level-set indicator has density or at . Therefore for every , and letting proves the limit is zero. Step 2.1 identifies this average with , proving norm differentiability of with derivative almost everywhere.
The norm of is exactly . The upper bound is step 1.1. For any partition with , take the step function with on each interval. It belongs to and satisfies , so . If , this lower bound is automatic. Taking the supremum over partitions gives . By [F2], , proving the displayed energy identity.
Depends on
- Bochner L2 on a finite interval is a separable Hilbert space
- Riesz representation for Hilbert spaces
- Bounded linear maps commute with Bochner integration
- Cauchy–Schwarz: $|\langle x,y\rangle|\le\|x\|\,\|y\|$, with equality exactly for dependent pairs
- Minkowski's inequality for integrals, including $p = \infty$
- The Axiom of Choice
- The Axiom of Countable Choice ($\mathrm{AC}_\omega$)
- The axiom of dependent choice: a relation in which every element is related to something admits an $\mathbb{N}$-indexed chain
- AC implies DC implies countable choice
- Bochner-integrable function
- Bochner integrability criterion
- Lebesgue differentiation theorem on $\mathbb{R}^n$
- Chebyshev-Markov inequality for the integral
- Finite and countable subadditivity of measures
Used by
- Derivative of squared W2 distance along a finite-action continuity equation Lemma
- Partial feature flows have exact energy and weight-moment identities Lemma
- The Hilbert lift induces the localized Wasserstein continuity equation Lemma
- A lower-bounded semiconvex Hilbert energy generates a unique gradient flow Theorem
Dependency tree · two levels
71 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
- Luigi Ambrosio, Nicola Gigli, and Giuseppe Savaré, Gradient Flows in Metric Spaces and in the Space of Probability Measures (standard reference, not scraped)
- Lénaïc Chizat and Francis Bach, On the Global Convergence of Gradient Descent for Over-parameterized Models using Optimal Transport (standard reference, not scraped)