Unsupervised Dictionary Learning for Anomaly Detection. (arXiv:2003.00293v2 [cs.LG] CROSS LISTED)

We investigate the possibilities of employing dictionary learning to address
the requirements of most anomaly detection applications, such as absence of
supervision, online formulations, low false positive rates. We present new
results of our recent semi-supervised online algorithm, TODDLeR, on a
anti-money laundering application. We also introduce a novel unsupervised
method of using the performance of the learning algorithm as indication of the
nature of the samples.

anomaly detection detection lg