Let's be real. Current neural nets have been shown empirically to work on some problems (after tinkering to get details right) - but do we really *know* that they are an abstraction of the brain, in which their details map onto simplifications of actual brain processes? No.
-
@GaryMarcus I think you're not understanding@tyrell_turing . I saw another post today that said "all neurology is inferior to ANN models," and THAT point of view is genuinely worth ridiculing.
only thing i was ridiculing (via parallel structure) was his certainty in the face of the unknown.
-
-
But my certainty is not the one which you were ridiculing. You kept trying to explain to me that ANNs might not capture the reality of the brain, and are not adequate models. But all I said was that they were an abstraction that captures some aspects of neural processing.
-
My point, again, is that we know so little about the overall nature of neural processing that is premature to say much more than X looks a little like Y that might do Z.
- 4 more replies
New conversation -
-
-
I think he was saying that ANN models might provide insight about specific brain modules / subsystems. I've seen some convincing papers specifically about cerebellum structure.
-
I don't think he was that certain; I think he was certain that we shouldn't disregard this tool, not that this tool is universal. But I can see how, to someone fed up with DL hype, what he said could strike a wrong note. My two cents. I'm out
- 1 more reply
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.