Jinwon An and Sungzoon Cho, “Variational autoencoder based anomaly detection using reconstruction probability,” Special lecture on IE, vol. 2, no. 1, pp. 1–18, 2015.
@article{an_variational_2015,
title = {Variational autoencoder based anomaly detection using reconstruction probability},
author = {An, Jinwon and Cho, Sungzoon},
journal = {Special lecture on IE},
volume = {2},
number = {1},
pages = {1--18},
year = {2015},
tldr = {In the context of anomaly detection with varational autoencoders, argues that reconstruction probability is a more objective measure than reconstruction error. Experiments with MNIST and KDD cup 1999 network intrustion dataset. The VAEs provide reconstructions as well as reconstruction probabilities.},
code = {https://paperswithcode.com/paper/variational-autoencoder-based-anomaly},
freepdf = {http://dm.snu.ac.kr/static/docs/TR/SNUDM-TR-2015-03.pdf}
}
We propose an anomaly detection method using the reconstruction probability from the variational autoencoder. The reconstruction probability is a probabilistic measure that takes into account the variability of the distribution of variables. The reconstruction probability has a theoretical background making it a more principled and objective anomaly score than the reconstruction error, which is used by autoencoder and principal components based anomaly detection methods. Experimental results show that the proposed method outperforms autoencoder based and principal components based methods. Utilizing the generative characteristics of the variational autoencoder enables deriving the reconstruction of the data to analyze the underlying cause of the anomaly.
tl;dr: In the context of anomaly detection with varational autoencoders, argues that reconstruction probability is a more objective measure than reconstruction error. Experiments with MNIST and KDD cup 1999 network intrustion dataset. The VAEs provide reconstructions as well as reconstruction probabilities.
Variational autoencoder based anomaly detection using reconstruction probability.
pdf: http://dm.snu.ac.kr/static/docs/TR/SNUDM-TR-2015-03.pdf
h/t Amy Tabbcode: https://paperswithcode.com/paper/variational-autoencoder-based-anomaly