First, LSTM is not Turing complete. I'll grant you that RNN architectures can be Turing complete, for instance ConvLSTM is Turing complete
Seen several replies to my post on the future of DL saying "who needs programming, LSTM is enough for everything (btw the brain is a LSTM)"
-
-
-
Second, what matters isn't that *in theory* you can express anything with a ConvLSTM. What matters is if you can do it in practice
- Show replies
New conversation -
-
-
Interesting thing if there's a continuum between LSTM-style and programming-style processing. E.g. methods to soften discreteness.
-
Basically what I suggest in my latest post: not generating source code, but something in between
End of conversation
New conversation -
-
-
Hmm. A human brain figures out how human brain works. Sounds like a fallacy
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You have to be outside of something to figure that something out
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.