Creating curious agents that can explore their environment and actions without obvious rewards seems like it'll lead to dramatic progress.
-
and quite legitimate in this case as the citations in the Pathak’s ICML paper reflect
It would be naive to attribute it to one computer scientist's 1991 paper, since the idea has been around in AI / neuro since the 1950s...
-
-
Important ideas typically take decades to mature in a field, they don't pop out of nowhere. It's incremental innovation all the way down.
-
It only *appears* as punctuated equilibrium because incremental innovation reaches a threshold and then explodes into a frenzy of progress.
End of conversation
New conversation -
-
-
I didn’t attribute anything. I just said someone did it in 1991
-
and the current paper takes it as an inspiration, among many other works
End of conversation
New conversation -
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.