Without wanting to "take sides" in the silly-yet-interesting debate/tweetfight between @GaryMarcus and @ylecun: when reading Gary's article it's worth keeping in mind that the dismissal of symbolic AI by many in deep learning is pretty much a reversal of 30 years ago.https://twitter.com/GaryMarcus/status/1068897657530138629 …
-
Show this thread
-
odd and perhaps mistaken claim. 30 years ago is precisely the last time time when proto-deep-learning advocates were writing papers about how (symbolic) linguistics was “alchemy”, just after the huge success in 1986 of the PDP volumes.
3 replies 0 retweets 5 likes -
Replying to @GaryMarcus @ylecun
I stand corrected - I was obviously not there. But was it not also the case that people from symbolic linguistics were initially very dismissive of neural net results?
1 reply 0 retweets 0 likes -
yes. (see last tweet) animus on both sides.
1 reply 0 retweets 0 likes -
Replying to @GaryMarcus @ylecun
You obviously know much more about these episodes than I do - I'm at least one generation removed from these events, and young DL people today at least two. I just felt that perspective was missing in your article, and could help explain some of the bitter feeling.
2 replies 0 retweets 1 like
fair enough. it was already quite bitter when i was applying to grad school in early 1989, dating at least to Perceptrons but probably to the 50s when Minsky/McCarthy vs Rosenblatt went different directions.
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.