deep learning papers - hassony2/inria-research-wiki GitHub Wiki
Papers
Dropout
Journal of Machine Learning Research 2014
[jmlr-dropout] Dropout: A Simple Way to Prevent Neural Networks from Overfitting [PDF] [notes]
Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov
Batchnorm
Journal of Machine Learning Research 2014
[jmlr-batchnorm] Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift [PDF] [notes]
Sergey Ioffe, Christian Szegedy
Autoencoders
Science 2006
science-autoencoder Reducing the Dimensionality of Data with Neural Networks [PDF] [notes]
Geoffrey Hinton, Ruslan Salakhutdinov
Notes
Convolutions
Dilated convolutions
about dilated convolutions and transposed convolutions