10-15 yrs ago, appears ~everybody talking cortical implementation of belief prop, ~nobody talking cortical implementation of back prop: https://scholar.google.com/scholar?q=related:CP4ffwsbGlcJ:scholar.google.com/&scioq=Cortical+circuitry+implementing+graphical+models&hl=en&as_sdt=0,5 … (not saying this is bad -- both very inspirational algorithms -- just interesting)
-
-
Replying to @AdamMarblestone
As someone said to me a few months ago: "Some people would argue backprop has taken off because of GPUs. If someone had made a GPU for Bayesian inference, then everybody would be flocking to that."
2 replies 0 retweets 5 likes -
Replying to @neurowitz
Vikash Mansingkha said *exactly* that to me a few years ago and I found it... disturbing.
2 replies 0 retweets 1 like
Replying to @AdamMarblestone @neurowitz
Look, I'm a big fan of *both* attempts to ground really naturally useful and real algorithms in potential biological implementations. In 2016, it felt somehow radical (especially to an outsider) to push backprop when the zeitgeist was still so probabilistic/Bayesian oriented...
4:34 PM - 14 Dec 2019
0 replies
0 retweets
1 like
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.