DeepFIB: Self-Imputation for Time Series Anomaly Detection

TL;DR

DeepFIB introduces a self-supervised imputation approach for time series anomaly detection, significantly improving accuracy by modeling anomalies as a fill-in-the-blank task and effectively handling different anomaly types.

Contribution

The paper proposes DeepFIB, a novel self-supervised learning method that enhances time series anomaly detection by using masking and imputation strategies to better capture temporal relations.

Findings

Outperforms state-of-the-art methods by up to 65.2% in F1-score

Effectively detects point- and sequence-outliers in time series data

Reduces anomaly detection errors with a new localization algorithm

Abstract

Time series (TS) anomaly detection (AD) plays an essential role in various applications, e.g., fraud detection in finance and healthcare monitoring. Due to the inherently unpredictable and highly varied nature of anomalies and the lack of anomaly labels in historical data, the AD problem is typically formulated as an unsupervised learning problem. The performance of existing solutions is often not satisfactory, especially in data-scarce scenarios. To tackle this problem, we propose a novel self-supervised learning technique for AD in time series, namely \emph{DeepFIB}. We model the problem as a \emph{Fill In the Blank} game by masking some elements in the TS and imputing them with the rest. Considering the two common anomaly shapes (point- or sequence-outliers) in TS data, we implement two masking strategies with many self-generated training samples. The corresponding self-imputation networks can extract more robust temporal relations than existing AD solutions and effectively facilitate identifying the two types of anomalies. For continuous outliers, we also propose an anomaly localization algorithm that dramatically reduces AD errors. Experiments on various real-world TS datasets demonstrate that DeepFIB outperforms state-of-the-art methods by a large margin, achieving up to $65.2\%$ relative improvement in F1-score.