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.
Show this thread
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.
-
-
-
Perhaps more importantly, although human engineers can cover a dramatically larger section of the space of interesting programs (compared to deep learning models), many useful programs remain inaccessible even to humans -- we cannot develop them by hand.
Show this thread -
It will be the job of future AI systems to give us access to these programs.
Show this thread
End of conversation
New conversation -
-
-
Very interesting to think about the kinds of programs only AI’s can make. I wonder what kind of programs these would be.
-
Mostly broken ones.
- Show replies
New conversation -
-
-
What are the processes that will be most efficient/effective at chewing through these data stacks.https://medium.com/@cleanapp/cryptos-killer-app-is-litter-ally-under-our-feet-1-of-5-eb064a6ab215?source=linkShare-7e3dc6459714-1529717135 …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
which article?
- Show replies
-
-
-
This Tweet is unavailable.
-
He wrote one of my favorite for programming with keras
End of conversation
-
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.