手法を比較

	制限付きボルツマンマシン (RBM)	オートエンコーダー	Variational Autoencoder
分野	深層学習	深層学習	深層学習
系統	Latent structure	Machine learning	Machine learning
提唱年	1986	2006	2014
提唱者	Smolensky, P. (1986); popularised by Hinton, G. E. & Salakhutdinov, R. R. (2006)	Hinton, G.E. & Salakhutdinov, R.R.	Kingma, D. P. & Welling, M.
種類	Generative energy-based probabilistic model	Neural network (encoder-decoder)	Deep generative latent-variable model (encoder–decoder)
原典	Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the Dimensionality of Data with Neural Networks. Science, 313(5786), 504–507.	Hinton, G.E. & Salakhutdinov, R.R. (2006). Reducing the Dimensionality of Data with Neural Networks. Science, 313(5786), 504–507.	Kingma, D. P. & Welling, M. (2014). Auto-Encoding Variational Bayes. International Conference on Learning Representations (ICLR).
別名	RBM, Harmonium, restricted Boltzmann machine, RBM generative model	Otokodlayıcı (Autoencoder), otokodlayıcı, auto-encoder, encoder-decoder network	Değişkensel Otokodlayıcı (VAE), VAE, auto-encoding variational Bayes, deep latent variable model
関連	3	4	5
概要	A Restricted Boltzmann Machine is a two-layer generative probabilistic model consisting of visible (observed) and hidden (latent) binary units connected by an undirected bipartite graph with no within-layer connections. Originally introduced as the 'Harmonium' by Paul Smolensky in 1986 and powerfully revived by Geoffrey Hinton and Ruslan Salakhutdinov in their landmark 2006 Science paper, RBMs became historically pivotal as the building block for greedy layer-wise pre-training of Deep Belief Networks, restarting interest in deep neural networks after years of stagnation.	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 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.