Metraipolis › archive › dc_32dfffd5003b
Bounding Retraining Equivalence and the Deletion Floor in Materials Machine Unlearning — arXiv:2609.35635
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35635
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:39:19.459Z
- SHA-256 of the archived text
eb4748c8cb2261482a6acb30f6cfe7c04b993a5259e323b27358b19a7a91646a- Stable URL of this page
https://metraipolis.com/documento/dc_32dfffd5003b
How to cite
Bounding Retraining Equivalence and the Deletion Floor in Materials Machine Unlearning — arXiv:2609.35635 — arXiv (en), https://arxiv.org/abs/2609.35635 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_32dfffd5003b (Metraipolis archive dc_32dfffd5003b, sha256 eb4748c8cb2261482a6acb30f6cfe7c04b993a5259e323b27358b19a7a91646a)
BibTeX
@misc{dc32dfffd5003b,
title = {Bounding Retraining Equivalence and the Deletion Floor in Materials Machine Unlearning — arXiv:2609.35635},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35635},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_32dfffd5003b; sha256 eb4748c8cb2261482a6acb30f6cfe7c04b993a5259e323b27358b19a7a91646a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35635. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In materials machine learning, closely related retained structures can sustain accurate property predictions even after removing a specific record, rendering post-deletion prediction error an ambiguous metric for machine unlearning. To resolve this ambiguity, we define the deletion floor as the expected target loss under a specified retraining procedure at the deleted request. Standard indistinguishability constraints yield a sharp interval bounding an update's target loss around this baseline reference. Theoretically, a conditional neighbor bound links a low deletion floor directly to retained fit, prediction regularity, and local label agreement, while an exact ridge identity isolates residual fit from the prediction change induced by record deletion. Empirically, controlled redundancy sweeps show an $\approx 8\times$ drop in median normalized retraining loss when one retained relative remains after deletion. Across two distinct fitting regimes in a paired Materials Project study, the lower-floor regime also exhibits a larger prediction change on more than 50% of the shared requests. Systematic comparisons against approximate updates and the original model decouple deliberate target suppression from preserved overall model utility. Consequently, request-level unlearning evaluations should report reference loss, prediction change, and retained utility together, interpreting post-deletion accuracy against what retraining itself leaves behind. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG) ; Materials Science (cond-mat.mtrl-sci); Quantum Physics (quant-ph). Pagina del lavoro: https://arxiv.org/abs/2609.35635 — 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 2026-10-07T22:14:28.495Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).