Metraipolis › archive › dc_b2cbc6515976

Learning to Unlearn: Machine Unlearning via Learning the Unlearning Behaviors — arXiv:2608.16700

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.16700
License (as archived)
CC0-1.0
Archived
2026-10-18T18:07:26.793Z
SHA-256 of the archived text
6e53a9c54e98e0a4136a4b2119e023863d36cf4a64ca61430aa3603bec58262a
Stable URL of this page
https://metraipolis.com/documento/dc_b2cbc6515976

How to cite

Learning to Unlearn: Machine Unlearning via Learning the Unlearning Behaviors — arXiv:2608.16700 — arXiv (en), https://arxiv.org/abs/2608.16700 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b2cbc6515976 (Metraipolis archive dc_b2cbc6515976, sha256 6e53a9c54e98e0a4136a4b2119e023863d36cf4a64ca61430aa3603bec58262a)
BibTeX
@misc{dcb2cbc6515976,
  title = {Learning to Unlearn: Machine Unlearning via Learning the Unlearning Behaviors — arXiv:2608.16700},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.16700},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b2cbc6515976; sha256 6e53a9c54e98e0a4136a4b2119e023863d36cf4a64ca61430aa3603bec58262a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16700. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Various machine unlearning techniques have been developed in response to privacy legislation requirements, enabling individuals to exercise their legal right to have their data $D_f$ removed from a machine learning model. This process is typically accomplished via the use of an unlearning function denoted as $U$. Existing methods focus on designing an intricate $U$ to unlearn $D_f \subset D$ from a previous model $A(D)$, so that the unlearned model performs as closely as possible to the retrained model $A(D \setminus D_f)$. However, these methods often suffer from high computational costs when dealing with massive training data, as the complex structures of $U$ become a bottleneck even for models with fewer parameters. Inspired by Learning to Optimize, we introduce the first learning-based model-agnostic approach, Learning-to-UnLearn (L2UL). Our core insight is to shift from manually designing $U$ to learning the unlearning behaviors from a distribution perspective, thereby acquiring a simple and efficient $U$ via learning. Our experimental results demonstrate that the accuracy achieved by L2UL is comparable to that of retraining while exhibiting impressive efficiency, particularly in data-intensive scenarios. Furthermore, we validate the performance and scalability of our method on larger models ResNet.

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/2608.16700 — 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 2026-10-18T16:29:14.141Z 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).