My perception is that in the of professional quantitative analysis, there remains some confusion—and perhaps some skepticism—about learning.

Extending these puzzles horizontally (by adding dimensions) and vertically (requiring certain puzzles to be solved before later combinations can be unlocked) might help to show people how to configure deep learning networks to tackle very complex problems.



Source link
thanks you RSS link
( https://www.reddit.com/r//comments/92ngju/r_deep__networks_and_the___/)

LEAVE A REPLY

Please enter your comment!
Please enter your name here