4 ms·
So if I'm reading this halfway correctly, quality isn't suddenly emergent, it's continuous and gradual based on size of the model. It only appears emergent when
by ianbutler 3y ago
So if I'm reading this halfway correctly, quality isn't suddenly emergent, it's continuous and gradual based on size of the model. It only appears emergent when researchers pick bad metrics.
I and I assumed a lot of people, already thought performance was a function on model size (# of parameters). Is this not what the prevailing thought is for DNN performance?
Agreed with the other posters that this title is misleading.
- freehorse 3y ago> I and I assumed a lot of people, already thought performance was a function on model size (# of parameters). I guess the disagreement has been in whether this function is "continuous" or not. I do not think the title is misleading, considering the article answers to quite specific claims in other articles. I agree it sounds misleading if you do not put it in that context.
- sgt101 3y agoI think most people expect(ed) that performance vs. size would (is) be an s-curve. The surprise for most is that we have climbed up the slopes so far and so fast. What the shape is is not clear to me.