English(EN) A Certified Robustness Framework for Backdoor Poisoning Attacks using $f$-Divergences

新框架认证机器学习模型对后门投毒攻击的鲁棒性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员开发了一个新的框架，用于认证机器学习模型在后门投毒攻击下的鲁棒性。该框架利用 $f$-散度来处理复杂的对手约束，包括被污染的训练样本数量和扰动幅度。它提供了一种通用的方法，可以将先前的方法作为特例恢复，并已用于认证模型在具有各种噪声分布的 $\ell_p$-约束投毒对手下的鲁棒性。 AI

影响增强了机器学习模型对抗对抗性攻击的安全性，可能增加对其部署的信任。

排序理由该条目是发表在arXiv上的研究论文，详细介绍了一个新的机器学习模型鲁棒性框架。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该条目是发表在arXiv上的研究论文，详细介绍了一个新的机器学习模型鲁棒性框架。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.LG TIER_1 English(EN) · Evan Rose, Alina Oprea · 2026-10-28 04:00

使用 $f$-散度针对后门投毒攻击的认证鲁棒性框架

arXiv:2610.34934v1 Announce Type: new Abstract: Machine learning systems trained on untrusted data are vulnerable to data and backdoor poisoning attacks, in which an adversary corrupts training and test samples to manipulate predictions at inference time. Randomized smoothing yie…