In cutting edge deployment of deep networks for different architectures (such as $CNN$, $QRNN$ etc) what is the historical trend of current limits of trainability possible computationally? By this I mean number of nodes and number of weights?

  • $\begingroup$ Hi. You need to define limits in a more precise way. $\endgroup$
    – nbro
    Dec 27 '19 at 12:36