Humans are born with millions of years of evolution. Since replicating that process in a literal way would be impractical, any kind of innateness we give AI won't be the sort that humans are born with.
-
-
-
this is like arguing that airplanes will never fly. surely we can reverse engineer some evolution’s endproducts and take hints from those in order to improve our systems.
- 3 more replies
New conversation -
-
-
When you design the architecture of an ANN you are instilling in it innate inductive biases that make it good at learning particular types of tasks. This may not be the form of innateness you would like to see, fine, but innate capabilities are not wholly ignored in DL.
-
Yes, I have made that point before.
End of conversation
New conversation -
-
-
Humans are born with a human body (which includes a learning brain). That is the most fundamental innateness. That body combined with the experience of a human existence within a human society gives rise to human-like intelligence. 1/2
-
chimps have somewhat similar bodies but wind up in a fairly different cognitive place. and humans with limited mobility still learn language. brain > body.
- 4 more replies
New conversation -
-
-
Innateness is bipedal, is a skeleton, a skull w/eyes gazing in a certain direction, organs that help to support action as in fight and flight and procreation, is all things language, hands with fingers to use a screw driver, type on a keyboard, is preparedness, genes, fitness...
- 1 more reply
New conversation -
-
-
Modem machine learning is an awesome typo. Trying to wrap my head around the concept.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
No, it's not the right kind, or at least the only kind, of innateness. If we knew the right kind of innateness we would be able to create human-level intelligence. At the moment we are experimenting with the kinds of innateness we can get to work, e.g. pre-trained ResNets etc.
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.