Ruslan Salakhutdinov
University of Toronto
1 paper in the Anchorcite directory, published in 2006, cited 21,380 times in total.
ORCID profileResearch topics
- Neural Networks and Applications · 1 paper
- Model Reduction and Neural Networks · 1 paper
- Image and Signal Denoising Methods · 1 paper
Frequent co-authors
- Geoffrey E. Hinton · 1 paper
Papers
- Reducing the Dimensionality of Data with Neural NetworksScience · 2006 · 21,380 citations
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…
Metadata from OpenAlex (CC0).