You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Just read the Berent & @GaryMarcus response. Summarizes familiar arguments against 1980s-style connectionism, but it would have been useful to see a discussion of more recent work, e.g.https://twitter.com/Joe_Pater/status/1048569045438541826 …
the old arguments still hold; we gave examples from Brendan Lakes work, a new paper from Pater on dropout (that i didn’t know of) makes clear that generalization to novel features remains outside the scope, two decades after i pointed out the problem
I agree that those are still good arguments, but your response seems to me to be too pessimistic about the odds of overcoming these issues in neural networks.
Perhaps neural networks with stronger inductive biases are not "real" networks anymore, but then it becomes an uninteresting terminological debate. I don't think as many people in the AI community these days really endorse the "associationist hypothesis" as you make it seem.
Marcus loves to point out problems that most of us have known about for a long time. His solution: let's return to the baby boomer symbolic AI of the last century and let's study CYC. Pathetic. 
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
