Time Series Anomaly Detection Using Convolutional Neural Networks and Transfer Learning

TL;DR

This paper introduces a CNN-based method with transfer learning for time series anomaly detection, pretraining on synthetic data and fine-tuning on real datasets, including a novel architecture for multivariate cases.

Contribution

It presents a novel CNN-based segmentation approach combined with transfer learning for anomaly detection in time series, including a new architecture for multivariate data.

Findings

Successful testing on synthetic and real datasets

Effective transfer learning from synthetic to real data

Novel architecture for multivariate time series

Abstract

Time series anomaly detection plays a critical role in automated monitoring systems. Most previous deep learning efforts related to time series anomaly detection were based on recurrent neural networks (RNN). In this paper, we propose a time series segmentation approach based on convolutional neural networks (CNN) for anomaly detection. Moreover, we propose a transfer learning framework that pretrains a model on a large-scale synthetic univariate time series data set and then fine-tunes its weights on small-scale, univariate or multivariate data sets with previously unseen classes of anomalies. For the multivariate case, we introduce a novel network architecture. The approach was tested on multiple synthetic and real data sets successfully.