Feathercrown @ Feathercrown @lemmy.world

Posts

6

Comments

4,125

Joined

2 yr. ago

I wouldn't consider "not being sent to a death camp" to be an extra protection that only applies to specific groups of people, though

But we're talking about a situation in which the protections are against unjust persecution. Selective lack of protections in this case is quite literally the same as selective persecution.

BREAKING: Water wet

Just get an even worse song stuck in your head, like this one: https://youtu.be/lrPncVUR_3Y

:D :D :D :D

5 guys burger with two patties, bacon, cheese, lettuce, tomato, grilled onion, and bbq sauce on a brioche bun (order number 44)

Nice catch. The carpet wear corroborates this.

This is beautiful

So based on that statistic, we should treat them differently? This line of thinking leads to some very bad places.

Don't lock people out of making valid arguments because they sound vaguely like arguments used by other people for negative means

Nobody should be subjected to this

He can be in it for himself, and it can still be a good thing for us

Am I stupid? Where's New Hampshire?

Undertale also has really bad code and it's a great game

Those are drawbacks

1. Waffle fries (always crispy and delicious)
2. Curly fries (same)
3. Tater tots (surprisingly good)
4. Wedges, if properly spiced
5. Good onion rings
6. Shoestring, if not limp
7. Bad onion rings
8. Bad wedges
9. Crinkle cut (zigzag) (usually unflavored and soggy from a bag, but can be better than soggy shoestrings)
10. Bad shoestrings
11. Sweet potato fries (always soggy and I don't like sweet potato very much)

So true actually

My point is that the way that you stick two real numbers together to make a complex number is important, and is unintuitive if you approach it as just two real numbers.

SOURCE: LAWS OF PHYSICS

Your “probably not” argument gets thinner every major AI update.

Right, but I'm talking about whether they're already using it, not whether they will in the future. It's certainly interesting to speculate about it though. I don't think we really know for sure how good it will get, and how fast.

Something interesting that's come up is scaling laws. Compute, dataset size, and parameters so far appear to create a limit to how low the error rate can go, regardless of the model's architecture. And dataset size and model size appear to require being scaled up in tandem to avoid over-/under-fitting. It's possible, although not guaranteed, that we're discovering fundamental laws about pattern recognition. Or maybe it's just an issue with our current approach.