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)"
-
Really it boils down to this. Finding & training a ConvLSTM that can render Markdown to HTML is near-impossible. But it's 20 lines of Pythonpic.twitter.com/ATke0YcDtC
Let's face it: we need richer ML primitives. We need more than matmul. I don't see why it's so difficult to grasp for some
-
-
Waiting to see if deep learning will benefit from the break throughs in understanding the huge role of glia in neuro. Neurons arent endgame
-
oh no, you just broke the deep learning hype. Now how will all those SV companies get funded at crazy AI valuations?
- Show replies
New conversation -
-
-
related, some devs implementing ML are purely content to treat it all as a black box. Interpretability is second class concern or less.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I like this post by David McAllester https://machinethoughts.wordpress.com/2016/06/20/cognitive-architectures/ … PS: LSTMs are Turing complete, no? with the right weight matrices?
-
Nevertheless, "Turing complete" means little "Beware of the Turing tar-pit in which everything is possible but nothing of interest is easy"
End of conversation
New conversation -
-
-
Do you have any primitive in mind that you think could be interesting?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
For me, that's been the strangest thing about learning ML. Hack after hack to work around not having more complex primitives
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.