Unsupervised Distance Metric Learning for Anomaly Detection Over Multivariate Time Series

arXiv:2403.01895v1 Announce Type: new
Abstract: Distance-based time series anomaly detection methods are prevalent due to their relative non-parametric nature and interpretability. However, the commonly used Euclidean distance is sensitive to noise. While existing works have explored dynamic time warping (DTW) for its robustness, they only support supervised tasks over multivariate time series (MTS), leaving a scarcity of unsupervised methods. In this work, we propose FCM-wDTW, an unsupervised distance metric learning method for anomaly detection over MTS, which encodes raw data …

abstract anomaly anomaly detection arxiv cs.ai cs.lg detection detection methods dynamic interpretability multivariate nature noise non-parametric parametric robustness series support tasks time series type unsupervised