An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series

TL;DR

This paper systematically evaluates deep learning methods for multivariate time series anomaly detection, introduces a new composite F-score metric, and finds that simple models with dynamic scoring outperform complex state-of-the-art techniques.

Contribution

It provides a comprehensive comparison of models and scoring functions, introduces a new evaluation metric, and identifies a simple auto-encoder with dynamic scoring as the best approach.

Findings

Dynamic scoring functions outperform static ones.

The composite F-score ($Fc_1$) effectively evaluates event detection.

A simple univariate auto-encoder with dynamic Gaussian scoring is highly effective.

Abstract

Several techniques for multivariate time series anomaly detection have been proposed recently, but a systematic comparison on a common set of datasets and metrics is lacking. This paper presents a systematic and comprehensive evaluation of unsupervised and semi-supervised deep-learning based methods for anomaly detection and diagnosis on multivariate time series data from cyberphysical systems. Unlike previous works, we vary the model and post-processing of model errors, i.e. the scoring functions independently of each other, through a grid of 10 models and 4 scoring functions, comparing these variants to state of the art methods. In time-series anomaly detection, detecting anomalous events is more important than detecting individual anomalous time-points. Through experiments, we find that the existing evaluation metrics either do not take events into account, or cannot distinguish between a good detector and trivial detectors, such as a random or an all-positive detector. We propose a new metric to overcome these drawbacks, namely, the composite F-score ($Fc_1$), for evaluating time-series anomaly detection. Our study highlights that dynamic scoring functions work much better than static ones for multivariate time series anomaly detection, and the choice of scoring functions often matters more than the choice of the underlying model. We also find that a simple, channel-wise model - the Univariate Fully-Connected Auto-Encoder, with the dynamic Gaussian scoring function emerges as a winning candidate for both anomaly detection and diagnosis, beating state of the art algorithms.