A New Interpretable Unsupervised Anomaly Detection Method Based on Residual Explanation

• Published in: IEEE access Vol. 10; pp. 1401 - 1409
• Main Authors: Oliveira, David F. N., Vismari, Lucio F., Nascimento, Alexandre M., de Almeida, Jorge R., Cugnasca, Paulo S., Camargo, Joao B., Almeida, Leandro, Gripp, Rafael, Neves, Marcelo
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Analytical models; Anomalies; Anomaly detection; Autoencoder; Catastrophic events; Computational modeling; Data models; Decision making; explainability; fault diagnosis; interpretability; Mathematical models; Predictive models; safety; Task analysis; Training

Despite the superior performance in modeling complex patterns to address challenging problems, the black-box nature of Deep Learning (DL) methods impose limitations to their application in real-world critical domains. The lack of a smooth manner for enabling human reasoning about the black-box decis...