Metraipolis › archive › dc_88bf16d39b52

Probabilistic Adversarial Training — arXiv:2609.39798

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39798
License (as archived)
CC0-1.0
Archived
2026-10-07T12:45:46.812Z
SHA-256 of the archived text
2d3452019d06721d71c4303a050dcb13589b7e5fa20a5ec1c6541e727294fd97
Stable URL of this page
https://metraipolis.com/documento/dc_88bf16d39b52

How to cite

Probabilistic Adversarial Training — arXiv:2609.39798 — arXiv (en), https://arxiv.org/abs/2609.39798 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_88bf16d39b52 (Metraipolis archive dc_88bf16d39b52, sha256 2d3452019d06721d71c4303a050dcb13589b7e5fa20a5ec1c6541e727294fd97)
BibTeX
@misc{dc88bf16d39b52,
  title = {Probabilistic Adversarial Training — arXiv:2609.39798},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39798},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_88bf16d39b52; sha256 2d3452019d06721d71c4303a050dcb13589b7e5fa20a5ec1c6541e727294fd97},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39798. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Building on a probabilistic perspective in which adversarial examples arise from the overlap between a distance-based distribution $p_{\mathrm{dis}}$ and a victim-classifier-induced distribution $p_{\mathrm{vic}}$, we start from a simple intuition: adversarial examples become harder to generate when these two distributions are pushed apart, as their overlap becomes smaller, thereby increasing robustness. This intuition naturally motivates a KL-based robustness objective. We then prove that $\mathrm{KL}(p_{\mathrm{dis}}\|p_{\mathrm{vic}})-\log Z_{\mathrm{vic}}$ is a lower bound on probabilistic robustness (PR), where $Z_{\mathrm{vic}}$ denotes the normalizing constant of $p_{\mathrm{vic}}$. Since PR is generally intractable to compute directly, maximizing this KL-based lower bound provides a tractable surrogate objective for improving PR. We further show that this objective recovers a scaled form of adversarial training, offering a probabilistic interpretation of adversarial training and a principled route to robustness improvement. We call the resulting method probabilistic adversarial training. Experiments show that it consistently improves PR, and ablation studies demonstrate that the induced scaling factor can even enhance the PR of non-probabilistic adversarial training methods.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2609.39798 — 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 2026-10-07T12:14:14.151Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).