Adaptive Memory Networks with Self-supervised Learning for Unsupervised Anomaly Detection

TL;DR

This paper introduces AMSL, a novel unsupervised anomaly detection method combining adaptive memory and self-supervised learning to improve generalization and robustness in detecting unseen anomalies, especially in large multivariate time series.

Contribution

It proposes AMSL, a new approach integrating adaptive memory and self-supervised learning within a convolutional autoencoder for enhanced anomaly detection.

Findings

Significantly outperforms state-of-the-art methods on four datasets.

Achieves over 4% improvement in accuracy and F1 score on a large sleep stage dataset.

Demonstrates increased robustness against input noise.

Abstract

Unsupervised anomaly detection aims to build models to effectively detect unseen anomalies by only training on the normal data. Although previous reconstruction-based methods have made fruitful progress, their generalization ability is limited due to two critical challenges. First, the training dataset only contains normal patterns, which limits the model generalization ability. Second, the feature representations learned by existing models often lack representativeness which hampers the ability to preserve the diversity of normal patterns. In this paper, we propose a novel approach called Adaptive Memory Network with Self-supervised Learning (AMSL) to address these challenges and enhance the generalization ability in unsupervised anomaly detection. Based on the convolutional autoencoder structure, AMSL incorporates a self-supervised learning module to learn general normal patterns and an adaptive memory fusion module to learn rich feature representations. Experiments on four public multivariate time series datasets demonstrate that AMSL significantly improves the performance compared to other state-of-the-art methods. Specifically, on the largest CAP sleep stage detection dataset with 900 million samples, AMSL outperforms the second-best baseline by \textbf{4}\%+ in both accuracy and F1 score. Apart from the enhanced generalization ability, AMSL is also more robust against input noise.