3 ms·
Talking about the benefits of dnn, what about the levels of abstraction? Each layer add levels of abstraction that you can't see in shallow networks.
by isseu 10y ago
Talking about the benefits of dnn, what about the levels of abstraction? Each layer add levels of abstraction that you can't see in shallow networks.
- clmcleod 10y agoI think the concept of "interpretability" is what you are getting at. I group that in with automatic feature engineering, since they are the same idea from different perspectives. Sometimes that is a benefit, sometimes it's not: http://blog.keras.io/how-convolutional-neural-networks-see-the-world.html http://blog.keras.io/how-convolutional-neural-networks-see-t...