After some dilly dallying, I finally decided to attend #BurningManAI er.. #NeurIPS2019--if only for a symbolic presence. Don't know if anyone else goes to this shindig--but if you are, and want to kibbitz, lemme know...
Symbolic presence ;) you are going to make Yann mad. He's out to destroy any mention of symbols 
@GaryMarcus
-
-
Well, y'know, part of the mission is disentangling the root cause from the outrageous...
#AIPeaceKeepingForcehttps://twitter.com/rao2z/status/1201313947472650240 … -
i did not and do not claim that Hinton started it; i mentioned Minsky and Papert’s role, pre-Hinton, and (perhaps in another tweet) noted that the war started at least as early as the 50s. TD was hardly only one puzzled; anyone who knows history of field knows YL was wrong.
- Show replies
New conversation -
-
-
Hey you forgot the Twitter etiquette rule that says: "if you are going to attack (
@ylecun), at least have the grace to tag". -
It's ironic that you talk about etiquette when I am talking about the precise lack of it. I am not attacking anyone. On the other hand I am precisely saying that personal attacks are unbecoming of a leader. I will block anyone who wishes to troll me on this.
End of conversation
New conversation -
-
-
I'm on record saying that one of the big challenges of AI is to make reasoning compatible with gradient-based learning. So I'm definitely not against "symbols". Many DL researchers are attempting to do just that. There is no war....
-
...if there is a war, it is from people who insist on a particular representation of symbols manipulated by discrete (non-differentiable) functions. My hunch, and Hinton's is: replace symbols by vectors of activities, and replace logic by smooth functions.
#NoWar - Show replies
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.