6 ms·
Here is Yann's summary of the situation back then: https://plus.google.com/+YannLeCunPhD/posts/gurGyczzsJ7 https://plus.google.com/+YannLeCunPhD/posts/gurGyczz
by kmiroslav 10y ago
Here is Yann's summary of the situation back then:
https://plus.google.com/+YannLeCunPhD/posts/gurGyczzsJ7 https://plus.google.com/+YannLeCunPhD/posts/gurGyczzsJ7
I seem to remember this article as having a lot of very interesting comments, including from the chairman of the conference. The exchange was cordial but showed some clear antagonism between the two parties. Somehow, I don't see these comments any more on Google+.
That was four years ago, things are a lot better now.
- rspeer 10y agoThings are better for neural nets because they've become fashionable. But nothing changed to make the peer review process tolerate unfashionable technologies.