Metraipolis › archive › dc_d8640dbedd84

Bifurcations from Weight Space: JEPAWG on Parametric Dynamical Systems — arXiv:2610.30847

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30847
License (as archived)
CC0-1.0
Archived
2026-10-23T15:15:08.695Z
SHA-256 of the archived text
4e1b611f6da520d3ca691342c9cab3fdff91677b23ceb9ef132d658a87394c45
Stable URL of this page
https://metraipolis.com/documento/dc_d8640dbedd84

How to cite

Bifurcations from Weight Space: JEPAWG on Parametric Dynamical Systems — arXiv:2610.30847 — arXiv (en), https://arxiv.org/abs/2610.30847 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d8640dbedd84 (Metraipolis archive dc_d8640dbedd84, sha256 4e1b611f6da520d3ca691342c9cab3fdff91677b23ceb9ef132d658a87394c45)
BibTeX
@misc{dcd8640dbedd84,
  title = {Bifurcations from Weight Space: JEPAWG on Parametric Dynamical Systems — arXiv:2610.30847},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30847},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d8640dbedd84; sha256 4e1b611f6da520d3ca691342c9cab3fdff91677b23ceb9ef132d658a87394c45},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30847. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Trained network weights are increasingly treated as data, but there is rarely ground truth about what a population of models should encode, which hampers systematic validation of weight-space methods. This makes parametric dynamical systems a natural test-bed: the dynamics change qualitatively at submanifolds of parameter space that are, in well-studied families, known analytically or numerically. Fine-tuning a forecasting network at each point of a parameter grid yields a model zoo indexed by a physical parameter, against which weight-space diagnostics can be checked. We build such model zoos for four dynamical systems and analyse them with JEPAWG, a recently proposed joint-embedding hypernetwork built around a learned parameter--weight representation. It was first introduced for neural samplers for lattice field theories, where its latent geometry located the phase transition and estimated the 2D Ising critical exponent. Applied to dynamical systems, JEPAWG is competitive on forecasting with a hypernetwork trained directly for that task, despite being supervised on the parameter--weight correspondence rather than on forecasts. Beyond forecasting, the norm of the parameter-derivative of the decoded weights locates Hopf points and curves, tracks a period-doubling cascade closely enough to give a qualitative estimate of the Feigenbaum constant, and detects a global bifurcation, which has no eigenvalue signature and is therefore invisible to linearisation. As such, JEPAWG provides interpretable insight into the dynamics that is not available from other hypernetworks. Together, these results indicate that JEPAWG is a workable instrument for reading physical structure off a population of trained networks.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Chaotic Dynamics (nlin.CD).
Pagina del lavoro: https://arxiv.org/abs/2610.30847 — 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 2026-10-23T15:09:08.559Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).