arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2503.05239cs.LGcs.AI
收藏

基于单一二元证书的鲁棒保形预测

Robust Conformal Prediction with a Single Binary Certificate

CISPA Helmholtz Center for Information Security(CISPA亥姆霍兹信息安全中心);University of Cologne(科隆大学)

Soroush H. Zargarbashi, Aleksandar Bojchevski

AI总结:

本文提出一种基于单一二元证书的鲁棒保形预测方法,通过阈值二值化样本,在仅需极少蒙特卡洛采样的情况下生成更小的鲁棒预测集,并消除了分数函数必须有界的限制。

摘要:

保形预测(CP)将任意模型的输出转换为预测集,并保证以(可调节的)高概率覆盖真实标签。鲁棒保形预测将这一保证扩展到最坏情况(对抗性)输入。现有基线方法通过对随机平滑的一致性分数进行界定来实现鲁棒性。在实践中,它们需要昂贵的蒙特卡洛(MC)采样(例如每个点约$10^4$个样本)来维持可接受的集合大小。我们提出一种鲁棒保形预测方法,即使在MC样本显著减少的情况下(例如CIFAR10上仅需150个样本)也能产生更小的集合。我们的方法通过一个可调节(或自动调节)的阈值对样本进行二值化,该阈值的选择旨在保持覆盖保证。值得注意的是,我们证明了仅需计算一个二元证书即可实现鲁棒性,而此前的方法需要为每个校准(或测试)点分别进行认证。因此,我们的方法速度更快,并能返回更小的鲁棒集合。我们还消除了先前方法中要求分数函数有界这一限制。

英文摘要:

Conformal prediction (CP) converts any model's output to prediction sets with a guarantee to cover the true label with (adjustable) high probability. Robust CP extends this guarantee to worst-case (adversarial) inputs. Existing baselines achieve robustness by bounding randomly smoothed conformity scores. In practice, they need expensive Monte-Carlo (MC) sampling (e.g. $\sim10^4$ samples per point) to maintain an acceptable set size. We propose a robust conformal prediction that produces smaller sets even with significantly lower MC samples (e.g. 150 for CIFAR10). Our approach binarizes samples with an adjustable (or automatically adjusted) threshold selected to preserve the coverage guarantee. Remarkably, we prove that robustness can be achieved by computing only one binary certificate, unlike previous methods that certify each calibration (or test) point. Thus, our method is faster and returns smaller robust sets. We also eliminate a previous limitation that requires a bounded score function.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑