Re:ARM Details “Project Trillium” Machine Learning Processor...
by chime on Friday June 01, 2018 @12:32AM
(#17202580)
It says the processor is built for efficient convolutions. While that would be great for CNNs, could the same be used to make RNNs more efficient? I.e. can the problem of solving RNNs be reduced to solving CNNs?
[ Parent
| on HN ]