Targeted collapse regularized autoencoder for anomaly detection: black hole at the center

arXiv:2306.12627v2 Announce Type: replace
Abstract: Autoencoders have been extensively used in the development of recent anomaly detection techniques. The premise of their application is based on the notion that after training the autoencoder on normal training data, anomalous inputs will exhibit a significant reconstruction error. Consequently, this enables a clear differentiation between normal and anomalous samples. In practice, however, it is observed that autoencoders can generalize beyond the normal class and achieve a small reconstruction error on some of the …

abstract anomaly anomaly detection application arxiv autoencoder autoencoders center cs.ai cs.cv cs.lg data detection development error inputs normal notion q-bio.nc stat.ml training training data type will