Quantile LSTM: A Robust LSTM for Anomaly Detection In Time Series Data

TL;DR

This paper introduces a robust LSTM-based anomaly detection method using quantile estimation and a new learnable activation function, demonstrating superior performance on industrial time-series datasets.

Contribution

It proposes a novel quantile-based anomaly detection approach combined with a Parametric Elliot Function activation in LSTM, enhancing anomaly detection accuracy.

Findings

LSTM-based quantile algorithms outperform existing methods

The new activation function improves long-range dependency modeling

Effective detection of anomalies in industrial datasets

Abstract

Anomalies refer to the departure of systems and devices from their normal behaviour in standard operating conditions. An anomaly in an industrial device can indicate an upcoming failure, often in the temporal direction. In this paper, we make two contributions: 1) we estimate conditional quantiles and consider three different ways to define anomalies based on the estimated quantiles. 2) we use a new learnable activation function in the popular Long Short Term Memory networks (LSTM) architecture to model temporal long-range dependency. In particular, we propose Parametric Elliot Function (PEF) as an activation function (AF) inside LSTM, which saturates lately compared to sigmoid and tanh. The proposed algorithms are compared with other well-known anomaly detection algorithms, such as Isolation Forest (iForest), Elliptic Envelope, Autoencoder, and modern Deep Learning models such as Deep Autoencoding Gaussian Mixture Model (DAGMM), Generative Adversarial Networks (GAN). The algorithms are evaluated in terms of various performance metrics, such as Precision and Recall. The algorithms have been tested on multiple industrial time-series datasets such as Yahoo, AWS, GE, and machine sensors. We have found that the LSTM-based quantile algorithms are very effective and outperformed the existing algorithms in identifying anomalies.