This kind of measured public discourse is exactly what the AI community needs more of, so as to mature as a scientific field. A strong culture of (constructive) criticism, what Popper called 'conjecture and refutation'. Let's make it happen!https://twitter.com/GaryMarcus/status/1187773364145836033 …
since so many people seem to miss your point, i would love some references; it's time I write about this. followed you so you can send references if you have them.
-
-
@DavidDeutschOxf has spoken at length about Popperian falsification and the importance of criticism (of ideas) in science. I'm sure there are many other references you could use (including Popper), but this one comes most prominently to mind.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
One thing to bear in mind is that Popper's perspective is probably too simplistic to be very useful for AI. The strictness of a falsification paradigm is not compatible with the development of highly complex theories that are partially correct.
@zerdeve is an expert on this -
I don’t think that our understanding of intelligence need consist of ‘highly complex theories that are partially correct’. The best theories are relatively simple & make predictions that can be tested. They ‘stick their neck out’.
- 1 more reply
New conversation -
-
-
I don't know we need to create more biological inspired tecnology. An deep learning system based on CNN, lstm and some rl or generative learning techniques is less intelligent the a rat.
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.