arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2512.04839hep-phhep-ex
收藏

用于异常检测的最优传输事件表示

Optimal Transport Event Representation for Anomaly Detection

Tongji University(同济大学);University of California, Berkeley(加州大学伯克利分校);Stanford University(斯坦福大学);SLAC National Accelerator Laboratory(SLAC国家加速器实验室)

Tianji Cai, Aditya Bhargava, Benjamin Nachman

AI总结:

提出最优传输(OT)作为弱监督异常检测的物理中间事件表示,在LHC Olympics基准数据集中仅注入0.5%共振信号时,OT增强特征集相比标准高观测量实现近两倍的显著性提升,而端到端深度学习在低信号区域效果较差。

摘要:

我们引入最优传输(OT)作为弱监督异常检测的基于物理的中间事件表示。在LHC Olympics基准数据集中仅注入0.5%的共振信号时,使用理想化设置,OT增强特征集相比标准高观测量实现了近两倍的显著性提升,而基于低层四动量的端到端深度学习在此低信号区域效果较差。观察到的增益在本研究考虑的信号类型和分类器中持续存在,表明结构化的、物理信息丰富的表示可以为现有异常检测方法提供有用的补充。

英文摘要:

We introduce optimal transport (OT) as a physics-based intermediate event representation for weakly supervised anomaly detection. With only $0.5\%$ injection of resonant signals in the LHC Olympics benchmark datasets, the OT-augmented feature set achieves nearly twice the significance improvement of the standard high-level observables using an idealized setup, while end-to-end deep learning on low-level four-momenta is less effective in this low-signal regime. The observed gains persist across signal types and classifiers considered in this study, suggesting that structured, physics-informed representations can provide a useful complement to existing approaches for anomaly detection.

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。

↑