Metraipolis › archive › dc_3d730666db41

Is Grokking a Loss of Normal Hyperbolicity of the Interpolation Manifold? — arXiv:2608.14803

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.14803
License (as archived)
CC0-1.0
Archived
2026-10-18T19:39:53.363Z
SHA-256 of the archived text
14e0060683bb7b0f1b5cd55107815459bd7d6fb333e22de73012bef9f8b52c07
Stable URL of this page
https://metraipolis.com/documento/dc_3d730666db41

How to cite

Is Grokking a Loss of Normal Hyperbolicity of the Interpolation Manifold? — arXiv:2608.14803 — arXiv (en), https://arxiv.org/abs/2608.14803 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3d730666db41 (Metraipolis archive dc_3d730666db41, sha256 14e0060683bb7b0f1b5cd55107815459bd7d6fb333e22de73012bef9f8b52c07)
BibTeX
@misc{dc3d730666db41,
  title = {Is Grokking a Loss of Normal Hyperbolicity of the Interpolation Manifold? — arXiv:2608.14803},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14803},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3d730666db41; sha256 14e0060683bb7b0f1b5cd55107815459bd7d6fb333e22de73012bef9f8b52c07},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14803. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A recent line of work recasts the post-memorization phase of grokking as constrained optimization: once a network interpolates the training set, weight decay drives a slow drift along the zero-loss manifold toward lower norm. In the language of dynamical systems, this is a fast-slow system in which the interpolation manifold plays the role of a slow manifold. We ask a question that this framing makes natural but the existing literature does not address: is the sharp generalization transition a loss of normal hyperbolicity of that manifold: a fold- or bifurcation-like event in which a normal restoring direction goes flat? Or does the manifold stay uniformly attracting while generalization happens by smooth drift? We propose a simple, optimizer-agnostic diagnostic: the smallest nonzero singular value $\sigma_{\min}^{+}(\mathbf J)$ of the residual Jacobian, which, for the squared loss, equals the slowest normal restoring rate of the manifold. On a two-layer ReLU network trained to grok modular addition under squared loss, $\sigma_{\min}^{+}(\mathbf J)$ does not collapse at the transition; it is near zero only before memorization and attains its largest values during the transition. The result holds across five seeds, and the six smallest singular values behave identically; there is no subspace-local collapse either. This is preliminary evidence against the bifurcation hypothesis and in favor of the smooth-contraction picture. We are explicit that a single-setting, gradual-transition experiment under Adam optimizer does not prove the absence of a bifurcation; it constrains where one could hide.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.14803 — 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 2026-10-18T19:02:44.056Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).