PT - JOURNAL ARTICLE
AU - Hinton, G. E.
AU - Salakhutdinov, R. R.
TI - Reducing the Dimensionality of Data with Neural Networks
AID - 10.1126/science.1127647
DP - 2006 Jul 28
TA - Science
PG - 504--507
VI - 313
IP - 5786
4099 - http://science.sciencemag.org/content/313/5786/504.short
4100 - http://science.sciencemag.org/content/313/5786/504.full
SO - Science2006 Jul 28; 313
AB - 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.