Porovnat metody

	Jádrová PCA	Autoencoder	Stroj s podpůrnými vektory (klasifikace)
Obor	Strojové učení	Hluboké učení	Strojové učení
Rodina	Latent structure	Machine learning	Machine learning
Rok vzniku	1998	2006	1995
Tvůrce	Schölkopf, B.; Smola, A. J.; Müller, K.-R.	Hinton, G.E. & Salakhutdinov, R.R.	Cortes, C. & Vapnik, V.
Typ	Nonlinear dimensionality reduction via kernel trick	Neural network (encoder-decoder)	Maximum-margin classifier (kernel method)
Původní zdroj	Schölkopf, B., Smola, A. J., & Müller, K.-R. (1998). Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation, 10(5), 1299–1319.	Hinton, G.E. & Salakhutdinov, R.R. (2006). Reducing the Dimensionality of Data with Neural Networks. Science, 313(5786), 504–507.	Cortes, C. & Vapnik, V. (1995). Support-Vector Networks. Machine Learning, 20, 273–297.
Další názvy	KPCA, kernel PCA, nonlinear PCA via kernel trick, kernel eigenvalue decomposition	Otokodlayıcı (Autoencoder), otokodlayıcı, auto-encoder, encoder-decoder network	Destek Vektör Makinesi (SVM — Sınıflandırma), support-vector network, SVM classifier, maximum-margin classifier
Příbuzné	5	4	5
Shrnutí	Kernel Principal Component Analysis (Kernel PCA) is a nonlinear dimensionality-reduction method introduced by Bernhard Schölkopf, Alexander Smola, and Klaus-Robert Müller in 1997–1998. It extends classical linear PCA to curved, non-linear data manifolds by implicitly mapping input data into a high-dimensional feature space via a kernel function, then performing standard PCA in that space — all without ever computing the mapping explicitly.	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.	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.