A Generative Process for Contractive Auto-Encoders.

6 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Better Mixing via Deep Representations (2012, methodology): “…paper to explore the preceding hypotheses are the Deep Belief Network or DBN (Hinton et al., 2006), trained by stacking Restricted Boltzmann Machines or RBMs, and the Contractive Auto-Encoder or CAE (Rifai et al., 2011a), for which a sampling algorithm was recently proposed (Rifai et al., 2012).”
Discriminative Unsupervised Feature Learning with Convolutional Neural Networks (2014, background): “Similarly, contractive autoencoders [10] penalize the sensitivity of the feature representation to small changes in the input.”
Multilingual Deep Learning (2013, background): “[18] propose a different regularization criteria which favours mappings that are more strongly contracting at the training samples.”
3D deep shape descriptor (2015, methodology): “Recently, deep models like deep auto-encoder [5, 48, 39], convolutional neural network [38, 29, 26], restricted Boltzmann machine [21, 33, 34] and their variants are widely used in computer vision applications.”