Deep learning is useful because it enables us to create programs that we could not otherwise code by hand. But the space of programs you can learn via deep learning models is a minuscule slice of the space of programs that we may be interested in.
-
Show this thread
-
Most interesting processes cannot be learned by a stack of affine transforms (plus squashing functions) mapping one vector space to another point-by-point. Even given large amounts of data.
7 replies 30 retweets 138 likesShow this thread
Replying to @alexiospanos
The space of possible programs is infinite, but we only care about a limited subset of them: programs that are useful to us.
3:26 PM - 22 Jun 2018
0 replies
0 retweets
2 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.