Ok, but: once you have come up with this architecture from first principles/theoretical considerations/biological inspiration, showing that it's not bullshit means actually building something that works, and so far nobody has come close.
AI and deep learning researchers, instead of trying to eke out another percentage point on latest perceptual benchmark, consider what architecture you might need to capture this, without extensive training data: New Caledonian crows create compound toolshttps://www.youtube.com/watch?v=xcN3WzeKDlE …
-
-
-
agreed, when it comes to general instellogence. deep learning or otherwise. that’s why we should be soul searching.
End of conversation
New conversation -
-
-
How much "training data" has a New Caledonian crow received over the course of its life, constantly interacting with objects and gaining a sense of physical dynamics? Less than an "extensive" amount?
-
The issue of course how to transfer from a lot of stuff that isn’t obviously task relevant (and also what prior knowledge might be leveraged).
End of conversation
New conversation -
-
-
Interesting example! Will take some time for RL agents (even curiosity-driven ones) to catch up ...https://twitter.com/awjuliani/status/970496935982333953?s=21 …
0:24 -
Some birds in Sendai, Japan, learned to use cars passing by the road to crack nuts:https://twitter.com/hardmaru/status/1063204983548432384?s=21 …
End of conversation
New conversation -
-
-
+1 Our blog is about AI systems architecture. Deep learning has it's place, but it's one box among many blocks. One size algorithm does not fit all problems!! Http://textician.com/blog
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.