Metraipolis › archive › dc_01e260fa2274
Beyond Gradient Flow: Identifiability and Recovery from Distribution Snapshots — arXiv:2609.35060
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35060
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:38:46.004Z
- SHA-256 of the archived text
f60054174dfe085443d0ed3e59939717f9bcd8edb66178e55e0a1f63394e18e9- Stable URL of this page
https://metraipolis.com/documento/dc_01e260fa2274
How to cite
Beyond Gradient Flow: Identifiability and Recovery from Distribution Snapshots — arXiv:2609.35060 — arXiv (en), https://arxiv.org/abs/2609.35060 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_01e260fa2274 (Metraipolis archive dc_01e260fa2274, sha256 f60054174dfe085443d0ed3e59939717f9bcd8edb66178e55e0a1f63394e18e9)
BibTeX
@misc{dc01e260fa2274,
title = {Beyond Gradient Flow: Identifiability and Recovery from Distribution Snapshots — arXiv:2609.35060},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35060},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_01e260fa2274; sha256 f60054174dfe085443d0ed3e59939717f9bcd8edb66178e55e0a1f63394e18e9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35060. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Inferring dynamics from snapshots of evolving distributions is fundamentally underdetermined: the Fokker-Planck equation constrains the drift $F$ only through its score-weighted divergence $\nabla\cdot F+F\cdot\nabla\log\rho$, leaving a $\rho$-solenoidal gauge invisible to any single-time constraint. Time-indexed transport formulations cannot resolve this ambiguity: every admissible marginal path admits a curl-free explanation, minimum-action reconstruction selects it, and marginal fit alone cannot distinguish dynamically inequivalent explanations. Requiring one autonomous field to explain several marginals instead makes part of the hidden circulation visible as $\nabla\log\rho$ changes across marginals. Separating instantaneous Fokker-Planck source constraints from the snapshot experiment, we show that the source constraints identify the field modulo the kernel of a stacked score-weighted divergence operator. For generic Gaussian shape variation, source constraints at $K\ge m$ time points in intrinsic dimension $m$ eliminate every polynomial gauge direction, whereas finitely many density snapshots alone admit aliasing; we give the obstruction explicitly. At a Gaussian anchor, for Sobolev smoothness $s$ and $n$ samples per time point, we derive a conditional lower rate $(nK)^{-2s/(2s+m+1)}$ for the tangent snapshot experiment, with a matching upper rate in a degreewise benchmark. Strong-form fitting is non-orthogonal to score error and cannot be repaired by spectral filtering. Instead, we estimate using smooth test functions while retaining the known diffusion term, and derive a finite-sample bound that separates sampling error from fixed-grid quadrature bias. Planted-circulation experiments confirm the predicted gauge contraction and expose a design tension between cross-slice information and covariance-aware whitening.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.35060 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T22:14:28.335Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).