方法对比

	自编码器	Isomap	支持向量机（分类）
领域	深度学习	机器学习	机器学习
方法族	Machine learning	Latent structure	Machine learning
起源年份	2006	2000	1995
提出者	Hinton, G.E. & Salakhutdinov, R.R.	Tenenbaum, J. B.; de Silva, V.; Langford, J. C.	Cortes, C. & Vapnik, V.
类型	Neural network (encoder-decoder)	Manifold learning / nonlinear dimensionality reduction	Maximum-margin classifier (kernel method)
开创性文献	Hinton, G.E. & Salakhutdinov, R.R. (2006). Reducing the Dimensionality of Data with Neural Networks. Science, 313(5786), 504–507.	Tenenbaum, J. B., de Silva, V. & Langford, J. C. (2000). A global geometric framework for nonlinear dimensionality reduction. Science, 290(5500), 2319–2323.	Cortes, C. & Vapnik, V. (1995). Support-Vector Networks. Machine Learning, 20, 273–297.
别名	Otokodlayıcı (Autoencoder), otokodlayıcı, auto-encoder, encoder-decoder network	Isomap, isometric feature mapping, geodesic Isomap, nonlinear MDS	Destek Vektör Makinesi (SVM — Sınıflandırma), support-vector network, SVM classifier, maximum-margin classifier
相关	4	3	5
摘要	An autoencoder is an encoder-decoder neural network, popularised by Hinton and Salakhutdinov in 2006, that compresses data into a low-dimensional latent code and then reconstructs it, enabling dimensionality reduction and anomaly detection. By learning to rebuild its own input through a narrow bottleneck, it discovers a compact representation of the data.	Isomap (Isometric Feature Mapping) is a manifold learning algorithm introduced by Tenenbaum, de Silva, and Langford in 2000 that discovers the intrinsic low-dimensional geometry of high-dimensional data by preserving geodesic — rather than straight-line Euclidean — distances between all pairs of points. It was one of the earliest, and most influential, nonlinear dimensionality reduction methods to demonstrate that genuinely curved data manifolds could be unfolded into a faithful low-dimensional coordinate system.	The Support Vector Machine, introduced by Corinna Cortes and Vladimir Vapnik in 1995, is a classifier that finds the optimal separating hyperplane between classes in a high-dimensional space. It chooses the boundary that leaves the widest possible margin to the nearest training points, which makes its decisions robust on new data.