Variational Inference for On-line Anomaly Detection in High-Dimensional Time Series

Abstract

Approximate variational inference has shown to be a powerful tool for modeling unknown complex probability distributions. Recent advances in the field allow us to learn probabilistic models of sequences that actively exploit spatial and temporal structure. We apply a Stochastic Recurrent Network (STORN) to learn robot time series data. Our evaluation demonstrates that we can robustly detect anomalies both off- and on-line.

Accepted as workshop paper at ICLR 2016; accepted as workshop paper for anomaly detection workshop at ICML 2016

Cited by in corpus (4)

• Deep Learning for Time Series Anomaly Detection: A Survey
• Unsupervised Time Series Outlier Detection with Diversity-Driven Convolutional Ensembles -- Extended Version
• Anomaly Detection in Multivariate Non-stationary Time Series for Automatic DBMS Diagnosis
• Batch Distillation Data for Developing Machine Learning Anomaly Detection Methods