Computer Science · cs.LG
Deep Anomaly Detection on Tennessee Eastman Process Data
Fabian Hartung, Billy Joe Franks, Tobias Michels, Dennis Wagner, Philipp Liznerski, Steffen Reithermann, Sophie Fellenz, Fabian Jirasek, Maja Rudolph, Daniel Neider, Heike Leitte, Chen Song, Benjamin Kloepper, Stephan Mandt, Michael Bortz, Jakob Burger, Hans Hasse, Marius Kloft
See on the mapAbstract
This paper provides the first comprehensive evaluation and analysis of modern (deep-learning) unsupervised anomaly detection methods for chemical process data. We focus on the Tennessee Eastman process dataset, which has been a standard litmus test to benchmark anomaly detection methods for nearly three decades. Our extensive study will facilitate choosing appropriate anomaly detection methods in industrial applications.
References
The reference list for this paper has not been collected yet.
Cited by 1
1 paper in the index cites this (Semantic Scholar counts 27 in total).
The cited-by list is still filling in while reference lists are collected.