Metraipolis › archive › dc_720b712faf94

SciCheck: A Framework for Verifying the Integrity of Deep Learning Training — arXiv:2611.09889

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09889
License (as archived)
CC0-1.0
Archived
2026-11-09T22:40:37.103Z
SHA-256 of the archived text
afbf84dd4a3c2a32bf9d14b4ab0247dbf8de387d6e2209e5dcd99d32e437682a
Stable URL of this page
https://metraipolis.com/documento/dc_720b712faf94

How to cite

SciCheck: A Framework for Verifying the Integrity of Deep Learning Training — arXiv:2611.09889 — arXiv (en), https://arxiv.org/abs/2611.09889 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_720b712faf94 (Metraipolis archive dc_720b712faf94, sha256 afbf84dd4a3c2a32bf9d14b4ab0247dbf8de387d6e2209e5dcd99d32e437682a)
BibTeX
@misc{dc720b712faf94,
  title = {SciCheck: A Framework for Verifying the Integrity of Deep Learning Training — arXiv:2611.09889},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09889},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_720b712faf94; sha256 afbf84dd4a3c2a32bf9d14b4ab0247dbf8de387d6e2209e5dcd99d32e437682a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09889. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deep learning training is difficult to verify within a scientific workflow: its architecture, hyperparameters, and procedure are rarely fixed until training has been attempted, and two honest runs never produce the same weights, so recomputation cannot serve as a check. We present SciCheck, a framework in which the scientist explores training configurations freely and posts a blueprint to a provenance ledger for each candidate they want verified, and a verifying scientist judges the resulting run in two layers: a statistical judge model that compares per-step gradient statistics against the verifier's own reference run, and a spot-check that re-executes randomly chosen training segments whose checkpoints are committed to the ledger one epoch at a time. On an MLP, a CNN, and an autoencoder at 100k-500k parameters, under gradient sign flipping, label poisoning, and fabricated runs, SciCheck reaches F1 of 0.966, 0.906, and 0.976 while rejecting 0-4.4% of honest runs. For the attacks we test, the spot-check accounts for most of this accuracy, and we characterize what each layer can detect. The effective-signal statistic falls as $(1-2s)^2$ under a sign flip of severity $s$, so gradient statistics alone catch strong enough sign flips, while fabricated runs that replay genuine statistics look no more anomalous than honest runs to any statistical detector we test. The spot-check catches fully fabricated segments with probability at least $1-(1-s)^k$ for $k$ sampled segments when the draws stay hidden until the run is committed, which roughly matches our measurements and lets a verifier choose $k$ for a target soundness. Verification costs 1.5-2.3x one retraining, mostly for the verifier's reference run; each submission's spot-check takes under 1.5 s, and per-epoch commitments reduce checkpoint storage 5-17x compared with keeping every checkpoint.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2611.09889 — 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 2026-11-09T22:33:17.041Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).