arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.09165cs.LGcs.AI
收藏

鲁棒且高效的共形预测集

Robust Yet Efficient Conformal Prediction Sets

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, Mohammad Sadegh Akhondzadeh, Aleksandar Bojchevski

AI总结:

针对共形预测易受规避与投毒攻击的问题,研究者通过界定一致性得分的最坏情况变化,推导出兼具可证明鲁棒性与更高效率的预测集,适用于多类数据与多种攻击场景。

摘要:

共形预测(CP)可将任意模型的输出转换为预测集,保证以用户指定的任意概率包含真实标签。但与模型本身一样,CP易受对抗性测试样本(规避攻击)和受扰动的校准数据(投毒攻击)的影响。我们通过界定一致性得分的最坏情况变化,推导出可证明鲁棒的预测集,更紧的边界带来了效率更高的集合。我们的方法覆盖连续和离散(稀疏)数据,且保证对规避和投毒攻击(针对特征和标签)均有效。

英文摘要:

Conformal prediction (CP) can convert any model's output into prediction sets guaranteed to include the true label with any user-specified probability. However, same as the model itself, CP is vulnerable to adversarial test examples (evasion) and perturbed calibration data (poisoning). We derive provably robust sets by bounding the worst-case change in conformity scores. Our tighter bounds lead to more efficient sets. We cover both continuous and discrete (sparse) data and our guarantees work both for evasion and poisoning attacks (on both features and labels).

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。

↑