다변량 시계열 이상 탐지 과업에서 비지도 학습 모델의 성능 비교
A Survey on Unsupervised Anomaly Detectionfor Multivariate Time Series
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
다변량 시계열 이상 탐지 과업에서 정답 값이 존재하는 데이터를 얻는 것은 매우 시간 집약적인 일이다. 따라서 최근 정답 값이 필요 없는 비지도 학습법(unsupervised learning)에 관한 많은 연구가 진행되었다. 하지만 다변량 시계열 이상 탐지 과업에 특화된 주요 구조와 세부적인 특성에 대한 심화 있는 논의는 이루어지지 않았다. 본 논문에서는 비지도 학습 기반의 다변량 시계열 이상 탐지 모델과 특장점을 포괄적으로 분석하여 분류하였다. 전력 계통(power grid) 또는 Cyber Physical System(CPS)과 같은 현실 세계 데이터 집합에서 현실적인 이상 상황을 고려하여 학습을 진행하였고, 실험 결과를 바탕으로 각 모델의 정량적 성능을 비교 분석하였다. 성능 지표로는 정밀도(precision), 재현율(recall)과 F1 점수를 사용하여 성능을 측정하였다.
Abstract
It is very time-intensive to obtain data with labels on anomaly detection tasks for multivariate time series. Therefore, several studies have been conducted on unsupervised learning that does not require any labels. However, a well-done integrative survey has not been conducted on in-depth discussion of learning architecture and property for multivariate time series anomaly detection. This study aims to explore the characteristic of well-known architectures in anomaly detection of multivariate time series. Additionally, architecture was categorized by using top-down and bottom-up approaches. In order to consider real-world anomaly detection situation, we trained models with dataset such as power grids or Cyber Physical Systems that contains realistic anomalies. From experimental results, we compared and analyzed the comprehensive performance of each architecture. Quantitative performance were measured using precision, recall, and F1 scores.
참고문헌 (21)
- B. Schölkopf, "Estimating the supportof a high-dimensional distribution", Neural computation, 13(7) (2001)
- K. Choi, "Deep learning for anomaly detectionin time-Series data : review, analysis, and guidelines", IEEE Access, 9, pp.120043~120065 (2021)
- C. Zhang, "Adeep neural network for unsupervised anomaly detection and diagnosis in multivariate time series data", Proceedings of the AAAI conference on artificial intelligence, 33(1), pp.1409~1416 (2019)
- X. Shi, "Convolutional lstm network : a machine learning approach for precipitation nowcasting", Advances in neural information processing systems, 28 (2015)
- J. Audibert, "Usad : unsupervised anomaly detection on multivariate time series", Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp.3395~3404 (2020)
- A. Shafahi, "Adversarial training for free!", Advances in Neural Information Processing Systems, 32 (2019)
- Y. Su, "Robust anomaly detection for multivariate time series through stochastic recurrent neural network", Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery &data mining, pp.2828~2837 (2019)
- J. Chung, "Empirical evaluation of gated recurrent neural networks on sequence modeling", NIPS 2014Workshop on Deep Learning (2014)
- D. Rezende, "Variational inference with normalizing flows", Proceedings of the32ndInternational Conference onMachineLearning, 37, pp.1530~1538 (2015)
- Z. Li, "Multivariate time series anomalydetection and interpretationusinghierarchical inter-metric and temporal embedding", Proceedings of the27thACM SIGKDD Conference onKnowledge Discovery &Data Mining, pp.3220~3230 (2021)
- L. Dinh, "Density estimationusingreal nvp", NIPS 2016 Deep LearningSymposium (2016)
- H. Zhao, "Multivariate time-series anomalydetection via graph attentionnetwork", IEEE International Conference on Data Mining(ICDM), pp.841~850 (2019)
- H. Ren, "Time-seriesanomaly detection service atmicrosoft", Proceedings of the 25thACM SIGKDD International Conference on Knowledge Discovery&Data Mining, pp.3009~3017 (2019)
- A. Deng, "Graphneural network-based anomaly detectioninmultivariate time series", Proceedingsof the AAAI Conference on Artificial Intelligence (2021)
- A. Rawashdeh, "Measuresof semantic similarity of nodes inasocial network", International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, 443, pp.76~85 (2014)
- A. Vaswani, "Attention is all you need", Advances in neural information processing systems, 30 (2017)
- J. Xu, "Anomaly transformer : time series anomaly detection with association discrepancy", Tenth International Conference on Learning Representations (2022)
- A. P. Mathur, "Swat : a water treatment testbed for research and training on ICS security", 2016 International Workshop on Cyber-physical Systems for Smart Water Networks, pp.31~36 (2016)
- C. M. Ahmed, "Wadi : a water distribution testbed for research in the design of secure cyber physical systems", 2017International Workshop on Cyber-physical Systems for Smart Water Networks, pp.25~28 (2017)
- D. Entekhabi, "Thesoil moisture active passive(smap)mission", Proceedings of the IEEE, 98(5), pp.704~716 (2010)
- J. P. Grotzinger, "Mars science laboratorymission and science investigation", Space Science Reviews, 170(1), pp.5~56 (2012)