I would like to cite Auto-Encoder based solution for anomaly detection, however I can't find the origin. When was Auto-Encoder used for anomaly detection for the first time?
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
A bit of reference chasing, combined with Google Scholar searches, suggests that the origin was
Japkowicz, N., Myers C., & Gluck M., (1995), “A Novelty Detection Approach to Classification”, in Mellish, C. (ed.) The International Joint Conference on Artificial Intelligence (IJCAI-95), Montreal, Canada. IJCAII & Morgan Kaufmann. San Mateo, CA. pp 518-523.
The paper is well-cited, demonstrates the utility of autoencoders for anomaly detection (see the CH46 helicopter gearbox fault detection example) and appears to have been treated as the key reference by others at the time. For example, this 1996 thesis states, "Recently, many studies have been released which use either auto-encoding networks or PCA to detect anomalies. The technique often used is based on [Japkowicz et al., 1995]".
