[HN Gopher] Google breaks the trillion-parameter ceiling with th...
___________________________________________________________________
Google breaks the trillion-parameter ceiling with the Switch
Transformer
Author : groar
Score : 19 points
Date : 2021-01-18 14:25 UTC (8 hours ago)
(HTM) web link (arxiv.org)
(TXT) w3m dump (arxiv.org)
| mensetmanusman wrote:
| It's an interesting thought experiment to consider models that
| have more parameters than there are data points being analyzed.
|
| What does that mean?
| rdlecler1 wrote:
| All networks are like this. In a network of N nodes you have
| N^2 potential relationships. That's just simple paired
| relationships. You can still go to higher order groups.
| igorkraw wrote:
| It basically means you need to learn about double descent:
|
| https://twitter.com/hippopedoid/status/1243229024085835779
|
| and regularize your models well, implicitly or explicitly. Your
| model might otherwise memorize the data instead of learning
| features.
___________________________________________________________________
(page generated 2021-01-18 23:03 UTC)