|
=== Anomaly detection ===
Another application for autoencoders is [[anomaly detection]].<ref>{{Cite journal |last=Morales-Forero, |first=A., & |last2=Bassetto, |first2=S. (|date=December 2019, December). |title=Case Study: A Semi-Supervised Methodology for Anomaly Detection and Diagnosis |url=https://ieeexplore.ieee.org/document/8978509/ In ''|journal=2019 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM)'' (p.|___location=Macao, 4)Macao (pp.|publisher=IEEE 1031-1037).|pages=1031–1037 IEEE|doi=10.1109/IEEM44572.2019.8978509 |isbn=978-1-7281-3804-6}}</ref><ref>Sakurada,{{Cite M.,journal &|last=Sakurada |first=Mayu |last2=Yairi, T. (2014,|first2=Takehisa |date=December). 2014 |title=Anomaly detectionDetection usingUsing autoencodersAutoencoders with nonlinearNonlinear dimensionalityDimensionality reductionReduction |url=http://dl.acm.org/citation.cfm?doid=2689746.2689747 In ''|journal=Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis - MLSDA''14 (p.|language=en 4).|___location=Gold Coast, Australia QLD, Australia |publisher=ACM Press |pages=4–11 |doi=10.1145/2689746.2689747 |isbn=978-1-4503-3159-3}}</ref><ref name=":8">An, J., & Cho, S. (2015). [http://dm.snu.ac.kr/static/docs/TR/SNUDM-TR-2015-03.pdf Variational autoencoderAutoencoder based anomalyAnomaly detectionDetection using reconstructionReconstruction probabilityProbability]. ''Special Lecture on IE'', ''2'', 1-18.</ref><ref>Zhou,{{Cite C.,journal &|last=Zhou |first=Chong |last2=Paffenroth, R.|first2=Randy C. (|date=2017, August).-08-04 |title=Anomaly detectionDetection with robustRobust deepDeep autoencodersAutoencoders |url=https://dl.acm.org/doi/10.1145/3097983.3098052 In ''|journal=Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining'' (pp.|language=en 665-674).|___location=Halifax NS Canada |publisher=ACM |pages=665–674 |doi=10.1145/3097983.3098052 |isbn=978-1-4503-4887-4}}</ref><ref>{{Cite journal|doi=10.1016/j.patrec.2017.07.016|title=A study of deep convolutional auto-encoders for anomaly detection in videos|year=2018|last1=Ribeiro|first1=Manassés|last2=Lazzaretti|first2=André Eugênio|last3=Lopes|first3=Heitor Silvério|journal=Pattern Recognition Letters|volume=105|pages=13–22|bibcode=2018PaReL.105...13R}}</ref> By learning to replicate the most salient features in the training data under some of the constraints described previously, the model is encouraged to learn to precisely reproduce the most frequently observed characteristics. When facing anomalies, the model should worsen its reconstruction performance. In most cases, only data with normal instances are used to train the autoencoder; in others, the frequency of anomalies is small compared to the observation set so that its contribution to the learned representation could be ignored. After training, the autoencoder will accurately reconstruct "normal" data, while failing to do so with unfamiliar anomalous data.<ref name=":8" /> Reconstruction error (the error between the original data and its low dimensional reconstruction) is used as an anomaly score to detect anomalies.<ref name=":8" />
Recent literature has however shown that certain autoencoding models can, counterintuitively, be very good at reconstructing anomalous examples and consequently not able to reliably perform anomaly detection.<ref>{{cite arXiv|last1=Nalisnick|first1=Eric|last2=Matsukawa|first2=Akihiro|last3=Teh|first3=Yee Whye|last4=Gorur|first4=Dilan|last5=Lakshminarayanan|first5=Balaji|date=2019-02-24|title=Do Deep Generative Models Know What They Don't Know?|class=stat.ML|eprint=1810.09136}}</ref><ref>{{Cite journal|last1=Xiao|first1=Zhisheng|last2=Yan|first2=Qing|last3=Amit|first3=Yali|date=2020|title=Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder|url=https://proceedings.neurips.cc/paper/2020/hash/eddea82ad2755b24c4e168c5fc2ebd40-Abstract.html|journal=Advances in Neural Information Processing Systems|language=en|volume=33|arxiv=2003.02977}}</ref>
|
