Metraipolis › archive › dc_db952f7f4ca0

Empirical Evaluation of Data Poisoning Attacks in Supervised Learning — arXiv:2609.10952

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.10952
License (as archived)
CC0-1.0
Archived
2026-10-16T05:46:20.934Z
SHA-256 of the archived text
c7adde08adaf6e7f0840086814723f047fdb49b578295714211d9d53f22d1a6a
Stable URL of this page
https://metraipolis.com/documento/dc_db952f7f4ca0

How to cite

Empirical Evaluation of Data Poisoning Attacks in Supervised Learning — arXiv:2609.10952 — arXiv (en), https://arxiv.org/abs/2609.10952 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_db952f7f4ca0 (Metraipolis archive dc_db952f7f4ca0, sha256 c7adde08adaf6e7f0840086814723f047fdb49b578295714211d9d53f22d1a6a)
BibTeX
@misc{dcdb952f7f4ca0,
  title = {Empirical Evaluation of Data Poisoning Attacks in Supervised Learning — arXiv:2609.10952},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.10952},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_db952f7f4ca0; sha256 c7adde08adaf6e7f0840086814723f047fdb49b578295714211d9d53f22d1a6a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10952. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Data poisoning corrupts training data to degrade a model or to plant attacker-controlled behavior. This study evaluates two representative training-time attacks, label flipping and backdoor poisoning, on MNIST and Fashion-MNIST with three baseline classifiers: Logistic Regression, Linear SVM, and Random Forest. Clean training is compared with poisoning rates of 5%, 10%, and 20% using clean-test accuracy, macro-precision, macro-recall, macro-F1, and, for backdoors, attack success rate. Label flipping caused clear degradation, largest for Logistic Regression and Linear SVM, while Random Forest stayed comparatively stable. Backdoor poisoning reached attack success rates from 0.9667 to 1.0000 on both datasets and all three models while often keeping clean-test performance near baseline. The results separate indiscriminate poisoning, which shows up in standard metrics, from targeted backdoor poisoning, which stays comparatively stealthy while embedding highly effective malicious behavior, and they support security-oriented evaluation beyond conventional clean-test metrics.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.10952 — 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 2026-10-16T05:30:39.380Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).