Metraipolis › archive › dc_91cc721229ca

Spectral Saliency for Machine Unlearning — arXiv:2608.15548

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.15548
License (as archived)
CC0-1.0
Archived
2026-10-18T18:03:34.683Z
SHA-256 of the archived text
9eb2bfd2c47724de0b1c9f28534cdda703e1475ebb4a9dc700dca9378af006f0
Stable URL of this page
https://metraipolis.com/documento/dc_91cc721229ca

How to cite

Spectral Saliency for Machine Unlearning — arXiv:2608.15548 — arXiv (en), https://arxiv.org/abs/2608.15548 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_91cc721229ca (Metraipolis archive dc_91cc721229ca, sha256 9eb2bfd2c47724de0b1c9f28534cdda703e1475ebb4a9dc700dca9378af006f0)
BibTeX
@misc{dc91cc721229ca,
  title = {Spectral Saliency for Machine Unlearning — arXiv:2608.15548},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15548},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_91cc721229ca; sha256 9eb2bfd2c47724de0b1c9f28534cdda703e1475ebb4a9dc700dca9378af006f0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15548. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine unlearning (MU) aims to remove the influence of specific training data while preserving model utility. As the name suggests, MU can be viewed as the inverse of learning, using gradient-based updates to reduce the influence of a forget-set by counteracting the previously learned behavior. Recently, Muon, a gradient descent variant, has been introduced. Muon applies spectral magnitude normalization to encourage exploration of rare directions and demonstrates promising performance. Inspired by Muon, we adopt the spectral view for unlearning and propose Spectral Saliency Unlearning (SSU). SSU thresholds weak singular components and updates only those directions supported by a confident unlearning signal. We further provide theoretical justification for this thresholding approach from the perspective of the forgetting-retention trade-off. Experiments across image classifiers, diffusion models, and LLMs demonstrate SSU's effectiveness.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.15548 — 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 2026-10-18T16:29:14.024Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).