mlsanityData to 14 Oct 2026

CAPMix: Robust KPI Anomaly Detection for AIOps in Noisy and Dynamic Environments

arXiv 2509.06419 ↗First posted 8 Sep 2025cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Time-series anomaly detection is crucial in AIOps for maintaining large-scale service reliability. In production, streams of Key Performance Indicators (KPI) are high-dimensional, non-stationary, and affected by noise, deployment changes, and latent anomalies, making real failures hard to distinguish from benign variation. Most existing methods assume either normality (learning from "normal" history) or rely on injected anomalies for training. Yet injected patterns often misalign with real failure modes, skewing decision boundaries -- aka. Anomaly Shift. We propose CAPMix, a controllable anomaly augmentation framework with prior-guided injection for realistic temporal behaviors. CAPMix combines label revision and dual-space mixup to enhance robustness under contaminated and mixed data. CAPMix consistently outperforms state-of-the-art methods on public AIOps and time-series benchmarks. It has been deployed in Kuaishou's large-scale production system, reducing false alarms and improving monitoring reliability. A real-world dataset is also released to enrich the research on robust KPI anomaly detection.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 30 of its 57 references that are in the map. First posted on arXiv 8 Sep 2025.

Builds on

30 of 57 references.

References · 30

  1. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  2. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  3. 2018Deep Anomaly Detection with Outlier Exposurecited by 942
  4. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  5. 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 410
  6. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 394
  7. 2021Time-Series Representation Learning via Temporal and Contextual Contrastingcited by 345
  8. 2019Deep Semi-Supervised Anomaly Detectioncited by 291
  9. 2018Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applicationscited by 224
  10. 2016LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detectioncited by 215
  11. 2019Time-Series Anomaly Detection Service at Microsoftcited by 165
  12. 2020Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progresscited by 151
18 more
  1. 2020Learning and Evaluating Representations for Deep One-class Classificationcited by 133
  2. 2024TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysiscited by 116
  3. 2024CATCH: Channel-Aware multivariate Time Series Anomaly Detection via Frequency Patchingcited by 94
  4. 2020Exathlon: A Benchmark for Explainable Anomaly Detection over Time Seriescited by 56
  5. 2020SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Datacited by 56
  6. 2021SmoothMix: Training Confidence-calibrated Smoothed Classifiers for Certified Robustnesscited by 49
  7. 2020Rethinking Assumptions in Deep Anomaly Detectioncited by 45
  8. 2021Neural Contextual Anomaly Detection for Time Seriescited by 43
  9. 2022Calibrated One-class Classification for Unsupervised Time Series Anomaly Detectioncited by 40
  10. 2023AnomalyBERT: Self-Supervised Transformer for Time Series Anomaly Detection using Data Degradation Schemecited by 25
  11. 2023Finding Order in Chaos: A Novel Data Augmentation Method for Time Series in Contrastive Learningcited by 18
  12. 2023Label-Free Multivariate Time Series Anomaly Detectioncited by 14
  13. 2024European Space Agency Benchmark for Anomaly Detection in Satellite Telemetrycited by 12
  14. 2022Deep Contrastive One-Class Time Series Anomaly Detectioncited by 10
  15. 2025Robust and Explainable Detector of Time Series Anomaly via Augmenting Multiclass Pseudo-Anomaliescited by 7
  16. 2023Beyond Sharing: Conflict-Aware Multivariate Time Series Anomaly Detectioncited by 5
  17. 2025GenIAS: Generator for Instantiating Anomalies in time Seriescited by 4
  18. 2025RoCA: Robust Contrastive One-class Time Series Anomaly Detection with Contaminated Datacited by 4

Built on by

Not cited yet.