.@ylecun This is outrageously false, and you know it. Rumelhart, Hinton and McClelland were attacking symbolists before I entered the field, in the 1986 PDP volumes, bitter at Minsky & Papert 1969.
and @ylecun if you are going to attack at least have the grace to tag me
1/5https://twitter.com/ylecun/status/1201233472989356032 …
-
Show this thread
-
Replying to @GaryMarcus @ylecun
I feel obligated to step in. I had John Hopfield's class in 1986. I went to the first IJCNN in San Diego in 1987. I *saw* it, especially the desperate and rude grant-seeking of both "sides." It's why I left the field. (1/4)
3 replies 0 retweets 2 likes -
Dan STAY THE F&$# HOME Greenberg Retweeted Dan STAY THE F&$# HOME Greenberg
In fact, I commented on it just a couple of days ago... https://twitter.com/titudeadjust/status/1200821304561164288 … So let me be the old, jaded guy here and comment on the factions based on decades of observation. (2/4)
Dan STAY THE F&$# HOME Greenberg added,
Dan STAY THE F&$# HOME Greenberg @titudeadjustReplying to @IntuitMachine @danbri @GaryMarcusI see this as partisan politics. Just like in national politics, there has been success of hyperpartisans in making noise to get the funding to the exclusion of centrist/hybrids/3rd parties. Saw it in spades in the 80s. Plus ca change...1 reply 0 retweets 0 likes -
.
@ylecun I was a NN purist in the '80's but have grown out of it. Arguing that Gary has not solved the problem of hybridization does not invalidate the argument that perceptron/NN/DL is probably insufficient alone for higher intelligence.4 replies 1 retweet 13 likes
The failure of symbolic AI and the present insufficiency of ANNs don't imply that we need hybrids other than for pragmatic reasons. Hybridization may not be the solution to bridge a false dichotomy. Our nervous systems are not ANNs, not symbolic, and not hybrids of both.
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.