0
I read in several review articles that deep leaning methods are divided into groups of Convolutional Networks, Boltzman Machines and Autoencoders. But none of these works justify why the division.
0No answers
Browser other questions tagged machine-learning deep-learning
You are not signed in. Login or sign up in order to post.
I think that would be the same as asking, why is programming divided into languages... It does not seem to me a division, but rather means for every purpose....
– MagicHat
These are just three neural network architectures, but they’re not the only ones. We have deep neural networks, recurring neural networks. Each architecture has a different purpose. Convolution works well with images, recurring networks with time series, autoencoders to generate an Encoder.
– AlexCiuffa
But if there are other architectures why Deep Learning is usually addressed to these architectures?
– Ewerton Duarte
Any @Magichat reference where I can confirm this?
– Ewerton Duarte
There is no reason to think about references, realize that through constant patterns groups are formed, and for each group there is a technology adapted exclusively for this group. When you understand that each group may contain a type of data, technologies are designed to optimize the ....
– MagicHat