What tutorial should I write next?
PS: why do you guys like autoencoders so much? Is it just because of Andrew Ng's ML class on Coursera, or is there something else?
-
-
@fchollet autoencoders are so 2014! ;) -
@seaandsailor@fchollet The dream of unsupervised learning is alive in {Theano, TensorFlow, Caffe, Torch, MXNet}. https://www.youtube.com/watch?v=TZt-pOc3moc … - Show replies
New conversation -
-
-
@fchollet don't slack off, write the next tutorial :)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet Can be useful as "auxiliary task" for RNNs when you not enough data / end-to-end signal is limited, i.e. http://arxiv.org/pdf/1508.05508v1.pdf … -
@fchollet Optimally you'd start from an RNN model trained on a "better" auxiliary task than autoencoding (language translation model etc...)
End of conversation
New conversation -
-
-
.
@fchollet I voted for autoencoders because I have never seen a practical use for them and I was hoping you could provide one.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet Theýre a simple example of auxil. tasks, which are so important in DL (I didn't vote 4 them thou ;-) )Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.