PeterisP

61 ms·

871.
by PeterisP 3y ago
At least for me (in EU) both PayPal and Revolut are literally banks and both provide deposit insurance - most likely because they don't have a choice.
872.
by PeterisP 3y ago
GDPR is likely not enforceable if you have no presence in EU whatsoever, if you have no assets in EU and no money coming in from EU. Anything Google does with data of EU residents is subject to GDPR even if that particular service is not of
873.
by PeterisP 3y ago
But for accuracy it doesn't matter if the relationship is causal, it matters whether the correlation is real. If in some country - for the sake of discussion, outside of Americas - a distinct ethnic group is heavily discriminated again
874.
by PeterisP 3y ago
No, the sampling does not work like that, that way lies madness (or poor results). The models oversample the most likely options and undersample rare options. Always picking the most likely option leads to bad outcomes, and literally sampli
875.
by PeterisP 3y ago
It's a big pain when trying to build things on top of the GPT-4 API. We had some experiments that were reliably, reproducibly achieving a goal, and then one day it suddenly stops working properly; then the student managed a different p
876.
by PeterisP 3y ago
I would distrust the currently available benchmarks, as recent research (gah, can't remember the paper title) indicates that for many benchmarks at least some of the data splits have leaked into model training data; and there's so
877.
by PeterisP 3y ago
Lawyers can and do make up bullshit briefs, but they're not allowed to do so, there is a requirement for them to ensure that they are valid - there is a clearly defined party who is responsible for any problems caused to the court, and
878.
by PeterisP 3y ago
Frankly, if people got a magic crystal ball and after hearing that warning in 1965 they would be able to know that there would be a 1.1C temperature rise by 2023 - in 58 years - then it still would make all sense for them to do nothing. For
879.
by PeterisP 3y ago
I think that the balance has swung so far due to how obnoxious and callous the adtech industry has become, that right now many people (including me) would consider it acceptable "collateral damage" if a bunch of decent sites would
880.
by PeterisP 3y ago
The adblocker settings can also be used to block various distracting content recommendation sections of sites you frequently use, if you want to just see the content you explicitly wanted without having your attention hijacked/tempted
881.
by PeterisP 3y ago
Because organizations rely on internal communication, and remote/on-site/hybrid each require very different processes to manage work and internal communications. If all your organization processes are set up to work well for on-si
882.
by PeterisP 3y ago
> That's not how it would be in a war. Their usage would be exactly zero, and then basically continuous. Not bursty at all - more like off, and then full on. IDK, I see it quite likely that even close to single-shot capability would
883.
by PeterisP 3y ago
No, debris in the same orbit would fall much more quickly than satellites, due to the square/cube difference, a larger piece gets slowed much less than many smaller pieces with the same combined mass. Also, "the same orbit" i
884.
by PeterisP 3y ago
If the launch capabilities are unbalanced (as they are!) then no, anti-satellite missiles aren't an option - at the moment noone can afford to launch an anti-satellite missile for every satellite that SpaceX can sustainably launch. Li
885.
by PeterisP 3y ago
An interesting "cheap, dirty & dangerous" mode of attack is loitering munitions from commodity FPV drones. Their capabilities have been demonstrated in current Russian invasion of Ukraine, and if any infiltrator can get in a f
886.
by PeterisP 3y ago
The big benefit of remote is being able to live 500 miles away and/or in a different country; and that requires being full-time remote. If he's anti full-time remote, then his pool of candidates is still limited to those who live
887.
by PeterisP 3y ago
AMD is not going down the path of ROCm; perhaps they claim to do so, but as evidenced by the lack of both effort and results, they clearly are not. The parent post is surprised that they still aren't making the appropriate investments
888.
by PeterisP 3y ago
One aspect of the specific positional embedding used there is that it explicitly encodes a signal that the very first attention layer can directly use for both relative and absolute position - i.e. that there can be a trivial set of wights
889.
by PeterisP 3y ago
Sure, the article doesn't say that something would be hanging around the moon permanently, it evaluates the damage that might occur if the orbiting spacecraft passes through the splash of debris ('the ejecta sheet') as it&#x2
890.
by PeterisP 3y ago
The way I see it, I'd expect that the generated comments would often get some human attention immediately afterward. Even if they don't get edited but there's just a crude "filter" where the developer keeps some com
891.
by PeterisP 3y ago
The paper models an 'ejecta sheet' of particles that exceed the escape velocity of the moon and thus aren't in orbit.
892.
by PeterisP 3y ago
I'd say that playing from notation and notating what's played are two very different things - in the latter, any difference between notation and execution means that it's the notation that is inaccurate; and in this particula
893.
by PeterisP 3y ago
The stats I can find show that 95%+ of Americans have less than $100k of savings, and much of savings is in non-cash means, so $250k per account seems completely sufficient to consider that yes, people are protected.
894.
by PeterisP 3y ago
It's a reasonable expectation that any and all institutions handling your money might be gangs of thieves run by charlatans. This is why any deviations from regulation should be viewed with extreme suspicion, since that's the on
895.
by PeterisP 3y ago
Well, no, the whole point is that a well edited cooking video would help a newbie to understand it, by explicitly and intentionally showing what exactly does "stir until golden brown" mean instead of just filming the whole cooki
896.
by PeterisP 3y ago
Did he do that consistently or is it just an artifact seen in a single very heavily edited record? Is the same timing seen in this section of live concert recordings during those years?
897.
by PeterisP 3y ago
It's not (solely) random imperfections - a good drum groove will have intentional specific deviations from the "theoretically exact" (bad) beat timing, no matter if done by a human or a properly engineered digital track. Ther
898.
by PeterisP 3y ago
How do you define "incorrectly"? IMHO the author of the song consistently plays it one way, then that's the correct execution of that rhythm, and playing it differently would be a different rhythm, not the one intended.
899.
by PeterisP 3y ago
Video editing is a thing. Something that takes 30 minutes to cook can (and should!) be cut down to 10 minutes of video. You cut five onions, you film how you cut one and edit out the rest. If you stir for 10 minutes watching for certain sig
900.
by PeterisP 3y ago
I'll just quote this same story: "An article in the Telegraph points to the added weight of batteries in EVs as the main risk factor, neatly sidestepping the fact that the average petrol-powered car is twice as heavy today as it w

More ›