arXiv cs.AI· Kadir-Kaan \"Ozer, Ren\'e Ebeling, Markus Enzweiler·· 15 小时前AI 评分24
ExCal：无需优化的多变量时间序列异常检测方法
ExCal: Optimization-Free Multivariate Time Series Anomaly Detection via Excess Calibration
AI 导读
ExCal 是一种仅用正常样本训练的多变量时间序列异常检测器，结合原始状态检索、信号/导数子空间与跨时间尺度稳健统计，无需梯度优化或异常标签。在 TSB-AD-M 的 17 个家族共 180 条序列上，ExCal 取得 AUC-PR 0.421±0.37 与 VUS-PR 0.452±0.37，在 24 种方法中排名第二，与 AxonAD、StreamVAE 无显著差异。
正文 · 原文
Abstract:We introduce ExCal, a normal-only multivariate time series anomaly detector that combines raw state retrieval, a signal/derivative subspace, and robust statistics across temporal scales. Each expert map receives a high normality floor from the training prefix. During scoring, positive excess is retained, attenuated for broadly firing maps, and combined with a small rank term. The method fits reference memories and projections without gradient optimization or anomaly labels. Prevalence and ranks depend on the complete scored sequence, making inference batch and transductive rather than online. On 180 series from 17 TSB-AD-M families, ExCal obtains AUC-PR $0.421\pm0.37$ and VUS-PR $0.452\pm0.37$ over three seeds, ranking second among 24 methods. Holm-adjusted paired tests detect no significant difference from AxonAD or StreamVAE, which are the two leading non-pretrained methods on TSB-AD-M, on either primary metric, while finding significant positive differences against the other 21 baselines. A separate single-run probe over three series measures median model-wrapper time of $0.0662$ s. Ablations associate the largest primary-metric changes with raw retrieval and excess-magnitude preservation. The evaluated score is decomposable by expert, but its diagnostic utility and online suitability remain untested.
| Comments: | Accepted at the 2026 IEEE International Conference on Big Data (IEEE BigData 2026), Special Session on Intelligent Data Mining. 10 pages, 3 figures, 9 tables |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.35568 [cs.LG] |
| (or arXiv:2610.35568v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35568 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kadir-Kaan Özer [view email]
[v1]
Mon, 26 Oct 2026 15:06:39 UTC (121 KB)
来源：arXiv cs.AI · arxiv.org