Reducing the dimensionality of data with neural networks. Hinton, G E & Salakhutdinov, R R Science, 313(5786):504–507, July, 2006.
Reducing the dimensionality of data with neural networks. [link]Paper  doi  abstract   bibtex   14 downloads  
High-dimensional data can be converted to low-dimensional codes by training a multilayer neural network with a small central layer to reconstruct high-dimensional input vectors. Gradient descent can be used for fine-tuning the weights in such "autoencoder" networks, but this works well only if the initial weights are close to a good solution. We describe an effective way of initializing the weights that allows deep autoencoder networks to learn low-dimensional codes that work much better than principal components analysis as a tool to reduce the dimensionality of data.
@article{hintonReducingDimensionalityData2006,
	title = {Reducing the dimensionality of data with neural networks.},
	volume = {313},
	url = {http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=16873662&retmode=ref&cmd=prlinks},
	doi = {10.1126/science.1127647},
	abstract = {High-dimensional data can be converted to low-dimensional codes by training a multilayer neural network with a small central layer to reconstruct high-dimensional input vectors. Gradient descent can be used for fine-tuning the weights in such "autoencoder" networks, but this works well only if the initial weights are close to a good solution. We describe an effective way of initializing the weights that allows deep autoencoder networks to learn low-dimensional codes that work much better than principal components analysis as a tool to reduce the dimensionality of data.},
	language = {English},
	number = {5786},
	journal = {Science},
	author = {Hinton, G E and Salakhutdinov, R R},
	month = jul,
	year = {2006},
	pmid = {16873662},
	pages = {504--507},
}

Downloads: 14