Metraipolis › archive › dc_0dc6c61664ab

Security Properties of Neural Networks as Decision Problems — arXiv:2609.39768

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39768
License (as archived)
CC0-1.0
Archived
2026-10-07T13:43:31.281Z
SHA-256 of the archived text
38b054d9052d8cbcc8ba931e565265e28dc41d70dca84a67241ae7a6f29f2dff
Stable URL of this page
https://metraipolis.com/documento/dc_0dc6c61664ab

How to cite

Security Properties of Neural Networks as Decision Problems — arXiv:2609.39768 — arXiv (en), https://arxiv.org/abs/2609.39768 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0dc6c61664ab (Metraipolis archive dc_0dc6c61664ab, sha256 38b054d9052d8cbcc8ba931e565265e28dc41d70dca84a67241ae7a6f29f2dff)
BibTeX
@misc{dc0dc6c61664ab,
  title = {Security Properties of Neural Networks as Decision Problems — arXiv:2609.39768},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39768},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0dc6c61664ab; sha256 38b054d9052d8cbcc8ba931e565265e28dc41d70dca84a67241ae7a6f29f2dff},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39768. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Certifying a deployed neural network raises decision problems that the verification literature has not classified: whether the model carries a backdoor planted in its training data, whether a fault in its stored parameters can drive it into an unsafe state, whether its output leaks a private part of its input. We formalise eight such problems and classify what we can. The organising observation is a logical one. The function computed by a piecewise linear network, together with all its node values, is definable by a quantifier-free formula of real addition of size linear in the network, so a property of the network is a quantifier-alternation sentence, which Sontag's 1985 theorem places in the polynomial hierarchy at the level of its prefix. Membership results are thus corollaries, and the argument makes plain what they need: that the quantified objects are inputs rather than the network's own parameters. Non-interference, monotonicity and counterfactual fairness have exactly the complexity of network equivalence and of interval verification, all co-NP- complete over ReLU. Detection of backdoor triggers from a quantised alphabet is Sigma_2^P-complete, one level above robustness certification, so it does not reduce to polynomially many robustness queries unless the hierarchy collapses. Inversion resistance is co-NP-complete for every l_p metric, p a fixed positive integer. Quantifying over parameters instead of inputs - the fault model of bit-flip attacks, radiation upsets and analog accelerators - makes verification exists-R-complete already for networks of identity nodes, for which every previously studied problem is in P, and it stays so when each parameter is confined to a box of inverse-polynomial width; the corresponding safety question is forall-R-complete for ReLU.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CR).
Categorie arXiv: Logic in Computer Science (cs.LO) ; Computational Complexity (cs.CC); Cryptography and Security (cs.CR); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.39768 — 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 2026-10-07T13:16:25.270Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).