arXiv：cs.AI· Seunghan Lee, Jun Seo, Jaehoon Lee, Junhyeok Kang, Sangjun Han, Sungdong Yoo, Minjae Kim, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Soonyoung Lee, Wonbin Ahn·· 16 小时前
OneShot：面向时间序列异常检测的标签高效检测器加权方法
OneShot: Label-Efficient Detector Weighting for Time Series Anomaly Detection
AI 导读
针对时间序列异常检测缺乏通用检测器的问题，研究者提出 ONESHOT，用单个已确认异常为整个检测器池加权，而非从中挑选单一模型。在 152 个单变量和 73 个多变量序列上，ONESHOT 优于最强单检测器、池平均，甚至优于在已确认事件上直接选择最佳检测器。代码已公开。
正文
Authors:Seunghan Lee, Jun Seo, Jaehoon Lee, Junhyeok Kang, Sangjun Han, Sungdong Yoo, Minjae Kim, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Soonyoung Lee, Wonbin Ahn
Abstract:Time series (TS) anomaly detection (AD) lacks a universal detector, with the best choice determined by the series rather than by the algorithm. Existing methods therefore either select one detector for each series or average a pool of them. However, both assume that no label ever arrives, while a deployed monitor does receive them, since an operator confirms the alarms it raises. In this paper, we propose ONESHOT, which uses a single confirmed anomaly to weight the pool rather than to choose one detector from it. Specifically, we score each detector on the confirmed anomaly, and those scores become the weights of a soft mixture over the whole pool. We show that a single confirmed anomaly is enough to come close to the best detector for the series. Across 152 univariate and 73 multivariate series, ONESHOT outperforms the strongest single detector, averaging the whole pool, and even picking the best detector on the confirmed event. Code is available at: this https URL.
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.16219 [cs.AI] |
| (or arXiv:2610.16219v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16219 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Seunghan Lee [view email]
[v1]
Mon, 12 Oct 2026 05:49:42 UTC (334 KB)
来源：arXiv：cs.AI · arxiv.org