Improved AutoEncoder With LSTM Module and KL Divergence for Anomaly Detection

The task of anomaly detection is to separate anomalous data from normal data in the dataset. Models such as deep Convolutional AutoEncoder (CAE) and deep support vector data description (SVDD) have been universally used and have demonstrated significant success in detecting anomalies. However, the o...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transactions on instrumentation and measurement Ročník 73; s. 1 - 11
Hlavní autori: Huang, Wei, Zhang, Bingyang, Zhang, Kaituo, Gao, Hua, Wan, Rongchun
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: New York IEEE 2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:
Accuracy
Anomalies
Anomaly detection
autoencoder
Data analysis
Data models
Datasets
deep support vector data description (Deep SVDD)
Feature extraction
hypersphere collapse
Long short term memory
LSTM
Mathematical models
Modules
Training
Vectors
ISSN:0018-9456, 1557-9662
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:The task of anomaly detection is to separate anomalous data from normal data in the dataset. Models such as deep Convolutional AutoEncoder (CAE) and deep support vector data description (SVDD) have been universally used and have demonstrated significant success in detecting anomalies. However, the over-reconstruction ability of CAE network for anomalous data can easily lead to high false-negative rate in detecting anomalous data. On the other hand, the deep support vector data description (Deep SVDD) model has the drawback of feature collapse, which leads to a decrease in detection accuracy for anomalies. To address these problems, we propose the Improved AutoEncoder with LSTM module and Kullback-Leibler divergence (IAE-LSTM-KL) model in this article. An LSTM network is added after the encoder to memorize feature representations of normal data. Meanwhile, the phenomenon of feature collapse can also be mitigated by penalizing the featured input to SVDD module via KL divergence. The efficacy of the IAE-LSTM-KL model is validated through experiments on both synthetic and real-world datasets. Experimental results show that IAE-LSTM-KL model yields higher detection accuracy for anomalies. In addition, it is also found that the IAE-LSTM-KL model demonstrates enhanced robustness to contaminated outliers in the dataset.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9456
1557-9662
DOI:10.1109/TIM.2024.3460931