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.
Wasserstein gradient flow for a feature-measure energy
Definition
Assume AC and the feature-measure energy hypotheses. Write for curves whose restriction to every finite interval belongs to in the metric. For with finite energy, define on , and extend this potential by off . Its constrained subdifferential at is the local Fréchet subdifferential of defined in the feature-measure-energy convention. A Wasserstein gradient flow from is a curve in , with , for which there is a Borel velocity satisfying the finite-action continuity equation and equals the negative minimum-norm element of for -almost every . This definition does not assert that the energy is displacement convex.
Depends on
Used by
- One-hidden-layer particle-measure model Definition
- Separated and nonseparated initial supports Example
- A homogeneous flow preserves spherical support separation Lemma
- A partial signed projection limit has convergent penalty and energy Lemma
- A weak two-homogeneous projection limit is stationary on its support Lemma
- Negative regular sublevels force two-homogeneous projection escape Lemma
- Nonoptimal partial projections have an escaping sign branch Lemma
- One-sided semiconvex characteristics give W2 uniqueness and stability Lemma
- Partial feature flows have exact energy and weight-moment identities Lemma
- Partial-homogeneity flows preserve box separation Lemma
- Stationarity of a partially homogeneous signed projection limit Lemma
- The Hilbert lift induces the localized Wasserstein continuity equation Lemma
- A partially homogeneous signed projection limit is a minimizer Theorem
- A separating two-homogeneous flow has globally minimizing projection limits Theorem
- The global Wasserstein gradient flow exists uniquely and is stable Theorem
- The localized Wasserstein gradient flow exists, satisfies EVI, and is unique Theorem
Dependency tree · two levels
26 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
- Lénaïc Chizat and Francis Bach, On the Global Convergence of Gradient Descent for Over-parameterized Models using Optimal Transport (standard reference, not scraped)
- Luigi Ambrosio, Gigli, and Giuseppe Savaré, Gradient Flows in Metric Spaces and in the Space of Probability Measures (standard reference, not scraped)