Hopfield Networks Is All You Need (2020)
40 points
4 days ago
| 4 comments
| arxiv.org
| HN
https://github.com/ml-jku/hopfield-layers
▲
jdefr89
1 day ago
[-]
So far I’ve needed hundreds of “all you needs” beginning to think papers with that title are not all I need… Let’s try and get a bit more creative than stealing the same name scheme used when Transformers were introduced.
reply
▲
IshKebab
16 hours ago
[-]
Tech authors seem to absolutely love hackneyed titles.

We're lucky it isn't "Falsehoods programmers believe about the unreasonable effectiveness of hopfield networks considered harmful, for fun and profit."

reply
▲
hodgehog11
1 day ago
[-]
I remember when this came out. It seemed cute, but appeared to have very little impact long-term unless I'm mistaken. Have there been any major developments from this?
reply
▲
dawnofdusk
1 day ago
[-]
Yann LeCun (Meta AI head) is still all-in on energy based models, which are basically Hopfield-like. Not sure if there are any major developments in that direction, but the hype/optimism is still there. https://iopscience.iop.org/article/10.1088/1742-5468/ad292b/...
reply
▲
dang
1 day ago
[-]
Related:

Hopfield Networks Is All You Need - https://news.ycombinator.com/item?id=26990000 - April 2021 (40 comments)

reply
▲
revskill
1 day ago
[-]
Gemini is good enough ?
reply