arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.08712cs.LG
收藏

Quantile LSTM:一种用于时间序列数据异常检测的鲁棒LSTM

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

Birla Institute of Technology and Science(比拉理工科学学院);MLSquare

Snehanshu Saha, Jyotirmoy Sarkar, Soma Dhavala, Santonu Sarkar, Preyank Mota

AI总结:

提出Quantile LSTM,通过估计条件分位数定义异常,并引入参数化Elliot函数作为LSTM激活函数建模长程依赖,在多个工业时间序列数据集上优于现有异常检测算法。

摘要:

异常是指系统和设备在标准操作条件下偏离其正常行为。工业设备中的异常通常在时间方向上预示着即将发生的故障。本文做出两项贡献:1)我们估计条件分位数,并考虑基于估计的分位数定义异常的三种不同方式。2)我们在流行的长短期记忆网络(LSTM)架构中使用一种新的可学习激活函数,以建模时间上的长程依赖性。具体而言,我们提出参数化Elliot函数(PEF)作为LSTM内部的激活函数(AF),与sigmoid和tanh相比,其饱和较晚。将所提出的算法与其他著名的异常检测算法(如孤立森林(iForest)、椭圆包络、自编码器)以及现代深度学习模型(如深度自编码高斯混合模型(DAGMM)、生成对抗网络(GAN))进行了比较。使用各种性能指标(如精确率和召回率)对算法进行评估。这些算法已在多个工业时间序列数据集(如Yahoo、AWS、GE和机器传感器)上进行了测试。我们发现基于LSTM的分位数算法非常有效,在识别异常方面优于现有算法。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑