anon291

3 ms·

1.
by anon291 5d ago
no amount of public money needs to be spent on companion dogs who can be cheaply euthanized.
2.
by anon291 5d ago
On the other hand, most states with significant tech workforces have Medicaid which most people will qualify for if they're laid off (I did!) and these are excellent. People pay for cobra despite a public option being available.
3.
by anon291 5d ago
I have never waited more than a day to see a doctor and hardly more than a week to see a specialist. I'm on a ppo plan so I just make an appt. If someone doesn't have an appointment I find someone else. The main issue I've ha
4.
by anon291 5d ago
That's a good question that can be answered by methods in mathematics.
5.
by anon291 5d ago
There is no evidence to substantiate what you are saying. Every possible proof exists already as a possible generation in the grammar of lean or rocq. In no way does that mean we have discovered everything.
6.
by anon291 9d ago
I mean if you're worried about high performance compute on cpus, then the argument also applies. That's why there so much hand written assembler with dynamic dispatch on CPU features... It's just that gpus have no general pur
7.
by anon291 9d ago
Because of how different the memory hierarchies and pipelines are and because if you're doing perf work, this is important enough to matter. For cpus and general purpose programs, it's not. For hpc, it is
8.
by anon291 9d ago
Defacing a photo Id(driver's license or passport) is technically a crime.
9.
by anon291 9d ago
Once again... There's literally no point to a low level shader language for heterogenous back ends.
10.
by anon291 10d ago
? I find it hard to see the issue here. Just put it in a separate file and call it?
11.
by anon291 10d ago
Wow that really flew over your head
12.
by anon291 10d ago
Right Europe famously has never meddled
13.
by anon291 10d ago
Yes I had an etrade account and made an investment. The schedule was extremely tight
14.
by anon291 11d ago
That's a decision for them to make. If they don't want it by all means don't force it. But to say you shouldn't talk to them is unfathomably cruel.
15.
by anon291 11d ago
I mean they keep banning people for random shit. I had a huge account, was even invited to the ipo due to my involvement I mentioned the imprisonment of Christians at seringapatam by Muslim rulers and I was immediately banned for islamophob
16.
by anon291 11d ago
It's not hard to make an old style building with modern amenities. I have a 1920s home and with like a few days of work it lives like a modern place. This is not some dichotomy.
17.
by anon291 11d ago
The brain doesn't need to do back propagation. Back propagation is a simulation of gradient descent which the universe achieved naturally by its own devices. Indeed if you tie the loss function to some sort of excited energy state in s
18.
by anon291 11d ago
It is immoral to treat another human being as too stupid to be exposed to others and worthy of your protection because you view them as primitive.
19.
by anon291 12d ago
some of them... lol. Not useful. A crap NVIDIA card can get you started with CUDA. You have to reach students. Without students, there's no future
20.
by anon291 12d ago
I hope his shareholders sue him to oblivion
21.
by anon291 12d ago
It's not talked about because the media frames AI as having agency on its own, which it does not. This talk from these companies of 'losing control' (no control was lost... Companies failed to monitor the workings of their ag
22.
by anon291 12d ago
China probably would win. If America regulates AI, you can bet the EU would do it ten times over
23.
by anon291 12d ago
Then the law is simple, no insurance? No product. No need to slow down development. Anyone who can make a good product should develop as fast as possible.
24.
by anon291 12d ago
English common law has been one of the greatest legal inventions of all time for business and accountability. Under common law a company that releases a product that has unintended consequences is liable for those. So naturally companies wa
25.
by anon291 12d ago
Which is kind of insane when you think about it considering what a cluster fuck the win32 API is
26.
by anon291 13d ago
I mean if you don't care about perf, opencl does what you want, and exists today. As for ai and matrix vector performance... I mean matrices are extremely useful across many domains. The hands off language that exists today is called
27.
by anon291 13d ago
It's not possible to do reliably. None of the models in current use today use any esoteric math. It's extremely easy to implement the math behind both the inference and learning of all modern models.
28.
by anon291 13d ago
amd consumer gpus don't support rocm. That is basically the entire reason for Nvidia dominance. All Nvidia cards support cuda even the crappy ones
29.
by anon291 13d ago
It's impossible to have an open standard. Hardware accelerators are nothing alike and have different perf characteristics. Each kernel is tuned to the hardware. The idea of writing a performant kernel in opencl is a fantasy. Source: wo
30.
by anon291 13d ago
That's literally not useful for the sort of applications people want to use transformers for.

More ›