Сравнение методов

	Автоэнкодер для обнаружения аномалий	Одноклассовая SVM	Вариационный автокодировщик
Область	Машинное обучение	Машинное обучение	Глубокое обучение
Семейство	Machine learning	Machine learning	Machine learning
Год появления	2006–2014	1999–2001	2014
Автор метода	Hinton, G. E. & Salakhutdinov, R. R. (autoencoders); applied to anomaly detection through multiple authors in the 2010s	Scholkopf, B., Platt, J. C., Smola, A. J., Williamson, R. C.	Kingma, D. P. & Welling, M.
Тип	Unsupervised deep learning (reconstruction-based)	Anomaly / novelty detection (unsupervised)	Deep generative latent-variable model (encoder–decoder)
Основополагающий источник	Chalapathy, R. & Chawla, S. (2019). Deep learning for anomaly detection: A survey. arXiv preprint arXiv:1901.03407.	Scholkopf, B., Platt, J. C., Shawe-Taylor, J., Smola, A. J., & Williamson, R. C. (2001). Estimating the support of a high-dimensional distribution. Neural Computation, 13(7), 1443–1471.	Kingma, D. P. & Welling, M. (2014). Auto-Encoding Variational Bayes. International Conference on Learning Representations (ICLR).
Другие названия	AE anomaly detection, reconstruction-error anomaly detection, deep autoencoder outlier detection, unsupervised autoencoder anomaly detection	OCSVM, one-class support vector machine, novelty SVM, unsupervised SVM	Değişkensel Otokodlayıcı (VAE), VAE, auto-encoding variational Bayes, deep latent variable model
Связанные	3	3	5
Сводка	Autoencoder anomaly detection trains a neural network to compress and then reconstruct normal data. Because the model has only ever learned what normal looks like, anomalous inputs produce noticeably higher reconstruction errors — and those errors become the anomaly score. The method requires no labeled anomalies and scales naturally to high-dimensional data such as sensor streams, images, and log records.	One-class SVM is an unsupervised anomaly and novelty detection algorithm that learns a tight boundary around normal training data in a kernel-induced feature space, flagging new observations that fall outside that boundary as outliers. Introduced by Scholkopf et al. in 1999–2001, it extends the SVM framework to the single-class setting where no labelled anomalies are available.	The Variational Autoencoder (VAE) is a deep generative latent-variable model, introduced by Diederik Kingma and Max Welling in 2014, that encodes data as a probability distribution in a latent space and samples from that distribution to generate new examples. It is used for data generation, anomaly detection, and feature learning.