Metraipolis › archive › dc_2da90ee053ff

SAM-on-the-Curve: Sharpness-Aware Mode Connectivity for Robust Weight-Space Interpolation — arXiv:2609.17748

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.17748
License (as archived): CC0-1.0
Archived: 2026-10-15T15:04:25.083Z
SHA-256 of the archived text: b372ad363ebb908c4a1e91038ca374201165d85963ea3695322fb20b4b010209
Stable URL of this page: https://metraipolis.com/documento/dc_2da90ee053ff

How to cite

SAM-on-the-Curve: Sharpness-Aware Mode Connectivity for Robust Weight-Space Interpolation — arXiv:2609.17748 — arXiv (en), https://arxiv.org/abs/2609.17748 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2da90ee053ff (Metraipolis archive dc_2da90ee053ff, sha256 b372ad363ebb908c4a1e91038ca374201165d85963ea3695322fb20b4b010209)

BibTeX

@misc{dc2da90ee053ff,
  title = {SAM-on-the-Curve: Sharpness-Aware Mode Connectivity for Robust Weight-Space Interpolation — arXiv:2609.17748},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17748},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2da90ee053ff; sha256 b372ad363ebb908c4a1e91038ca374201165d85963ea3695322fb20b4b010209},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17748. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deep neural networks that are independently trained to similar performance can be connected by low-loss parametric curves in weight space, a phenomenon known as Mode Connectivity (MC). This geometric property underpins practical techniques such as weight averaging, model ensembling, and model merging. We argue that low-loss connectivity is an incomplete geometric criterion: it controls loss only along a one-dimensional trajectory while leaving the surrounding weight-space neighborhood unconstrained, so the optimized curve may traverse sharp ridges that become fragile under distribution shift. We therefore reformulate mode connectivity as a neighborhood-robust path optimization problem, seeking a curve whose entire local neighborhood maintains low loss. We propose Sharp Mode Connectivity (SMC), which applies a first-order sharpness-aware approximation to the resulting minimax functional, enforcing flatness along the entire curve rather than only on it. We derive a practical optimization algorithm for connectivity paths under this sharpness-aware objective. Under severe blur corruptions from CIFAR-10-C, SMC achieves up to 6.09\% absolute accuracy improvement over standard MC. Remarkably, SMC produces negative loss barriers, meaning that models obtained at interior points of the optimized path can outperform the average endpoint loss. These results, validated across ResNet-18, VGG16-BN, and ViT-Tiny on CIFAR-10 and ImageNet-100, establish path-wise flatness as a practical principle for robust weight-space interpolation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.17748 — 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 2026-10-15T14:08:19.064Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).