A good deep learning architecture is one that introduces correct priors about the problem at hand -- in particular, about the structure of the correlations found in the data.
Show this thread
For instance, convolution is preferable to dense layers for data dimensions that are locally autocorrelated and translation-invariant. And depthwise separable convolutions are superior to convolution for any data that is locally autocorrelated and where channels are decorrelated.
-
-
(Which is naturally the case past the first layer of any convnet.)
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What is a channel?
-
With respect to a color images, typically the separate Red, Green and Blue channels.
End of conversation
New conversation -
-
-
What are some good examples of translation-invariant data dimensions? The rest makes sense BTW.
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.