Both ANNs and brains are (1) highly distributed and parallel (2) make small changes (3) efficient. Ideas like generalization hold for all systems with these three properties. In that sense, @tyrell_turing is obviously right. Maybe @GaryMarcus uses narrow def of abstraction.https://twitter.com/GaryMarcus/status/1200088188934451200 …
-
Proof is trivial if you use the proper units. Energy, then the difference is by orders of magnitude. Also, note, there are children (say with hyperlexia) that can (mostly) teach themselves to read by age 2-4.
there are AI systems that can teach themselves to read by age 12 or 14. oh wait. no, there aren’t. no AI matches human brain in ultimate attainment in reading comprehension. isn’t that a measure of model inadequacy?
-
-
The comparison is also unfair in that the child is not just focused on a reading task but also learning speech, visual parsing, sounds, language, theory of mind etc., all at the same time (and has spent much of their early years asleep).
-
OK sure. There are multiple things going on. But how many bits of information is the human child receiving per waking hour? You’ve avoided my question. And I suspect humans consume quite a bit of information before they do “one shot” feats.
- 22 more replies
New conversation -
-
-
I feel that you have made this point about a million times already. I don't think saying it one more time is going to accomplish anything that you might want.
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.