Metraipolis › archive › dc_8213805dca08

Why Forget-Only Unlearning Needs Memorization — arXiv:2610.10519

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10519
License (as archived)
CC0-1.0
Archived
2026-10-08T12:36:32.712Z
SHA-256 of the archived text
e9501235f041bd75866de5df8e4589635bf29e100f72b0d6486b8aa075d6514d
Stable URL of this page
https://metraipolis.com/documento/dc_8213805dca08

How to cite

Why Forget-Only Unlearning Needs Memorization — arXiv:2610.10519 — arXiv (en), https://arxiv.org/abs/2610.10519 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8213805dca08 (Metraipolis archive dc_8213805dca08, sha256 e9501235f041bd75866de5df8e4589635bf29e100f72b0d6486b8aa075d6514d)
BibTeX
@misc{dc8213805dca08,
  title = {Why Forget-Only Unlearning Needs Memorization — arXiv:2610.10519},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10519},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8213805dca08; sha256 e9501235f041bd75866de5df8e4589635bf29e100f72b0d6486b8aa075d6514d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10519. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine unlearning asks for a deletion algorithm whose output is close to retraining from scratch without the selected forget examples. In this work, we study forget-only unlearning, where the deletion algorithm receives only the trained model and the examples to forget, with no retained data or extra training information. We ask whether forget-only unlearning is always possible. We first show that this depends on the learning method: different datasets can produce the same trained model but require very different outputs after the same examples are removed. Using this observation, we derive lower bounds on how accurately unlearning can match retraining and instantiate them for several standard learning algorithms. We then ask what must be true when forget-only unlearning succeeds. To this end, we derive lower bounds on what an algorithm must memorize about the training data to handle arbitrary deletion requests. For simple threshold learners, the required information can be as large as the entire dataset, even though ordinary training keeps only one boundary point. Overall, our results show that information discarded during ordinary learning may be needed later for deletion, so models designed for forget-only unlearning may need to retain more information than standard training does.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Information Theory (cs.IT); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.10519 — 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 2026-10-08T12:14:15.354Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).