autoencoder

AutoEncoder (AE)

citekey	SakuradaYairi2014Anomaly
source code	own
Learning type	semi-supervised
input	multivariate

Notes

• AutoEncoder is trained on all non-anomaly data. Whenever it encounters an anomaly value, the reproduction error is quite higher than the error with non-anomaly instances.
• The test data is put into the AutoEncoder and the scores are returned.
• An assumption is made that all errors are normally distributed with some mean and std. Any error value that follows mean + kstd > threshold or mean - kstd < thereshold is considered as an anomaly. (NOT IMPLEMENTED)