cochne

7 ms·

1.
by cochne 2mo ago
Are you sure it wouldn’t suggest such a thing? Stoic thinkers (such as the one your repo is named after) are not necessarily anti-suicide. And I think if you prompted a real stoic with the question and the right context, it would certainl
2.
by cochne 4mo ago
The public “funded” these models in the sense of contributing to their training data.
3.
by cochne 1y ago
Wouldn’t it be 1/(number of days since 1888) for a given day? Or 1/(number of years) for a given year? So less than 1/100 if static and independent
4.
by cochne 1y ago
I started using an AI calorie tracking app. For non-branded foods, the calorie counting is pretty inaccurate and sometimes very off. However, having an app that can track pictures and create short descriptions that someone else (my traine
5.
by cochne 2y ago
I'm in your boat. I tried out TikTok out a few times, including making a new account, but it never showed me good content. I had maybe one or two longer sessions, but never felt the need to go back, like I (unfortunately) do with Red
6.
by cochne 2y ago
There is no thresholding, you delete edges randomly based on the probability assigned.
7.
by cochne 2y ago
>It gives me hope that teenagers are watching his videos and becoming inspired to go into infrastructure. Sadly, that was me ~10 years ago, but the lure of FAANG money was too strong and I went into EE/CS after 1 year as a civil eng
8.
by cochne 3y ago
I think they are correct, do you have a source? From my knowledge the only other components are the fully connected networks which are not big contributors.
9.
by cochne 3y ago
I hardly think it's fair to say that they were 'taking advantage'. If Google says unlimited, should the typical person really expect that to be taken away? That's a really bad look for Google. "If we offer someth
10.
by cochne 3y ago
I never really got how proofs are supposed to solve this issue. I think that would just move the bugs from the code into the proof definition. Your code may do what the proof says, but how do you know what the proof says is what you actua
11.
by cochne 3y ago
NP hard problems aren’t all equivalent, that’s NP complete.
12.
by cochne 3y ago
I don’t think that would necessarily favor small artists. That would just favor artists who are listened to by people who don’t use Spotify a lot. Right now someone who only streams a few songs gets a very small “vote” (assuming pay is per
13.
by cochne 3y ago
Another way to look at it is that you are now paying less for the base ticket which used to include those things. This allows consumers more choice in what part of the experience they want to pay for. If someone wants to build a computer a
14.
by cochne 3y ago
For those who are unaware 'engineering' in Swiss German is "Ingenieurwissenschaften"
15.
by cochne 3y ago
The thing about Kelly is that it assumes risk of ruin is zero because you never bet your entire bankroll. It assumes you can always bet a fraction of a fraction of what you have remaining.
16.
by cochne 3y ago
You can certainly write a program which can determine if a program would halt. It just won't be able to output a yes or no for _every_ program. In some cases it would have to output 'undecidable'. Most of the practical pro
17.
by cochne 3y ago
> Not every piece of code is an abstraction of course. To me, an abstraction is a piece of code that’s expressed in high-level language so that the distracting details are abstracted away. If I were to see a confusing piece of code litte
18.
by cochne 3y ago
So people polluting the atmosphere will actually have to pay for that? Oh the humanity!
19.
by cochne 3y ago
> I'm new to the codebase and I just want to know how it (the codebase) works. You should want to know _why_ it works that way too if you want to do any meaningful work with it. Context matters. I’ve seen many cases where the way
20.
by cochne 3y ago
I can’t say I know the truth, but the sentiment is that Welch divested from R&D, focused on “financialization” I.e. investing in financial instruments over adding real value, made big short term gains, and doomed the future of the compa
21.
by cochne 3y ago
What is unscientific about the article?
22.
by cochne 3y ago
What substantial features have they been pushing? To me it looks like just a bunch of ways to limit and monetize their existing userbase.
23.
by cochne 3y ago
What precisely is your criticism of the post?
24.
The Thoughtful Execution Framework(spotify.design)
2 points by cochne 3y ago|0 comments
25.
by cochne 3y ago
Yep. It is now my policy to not do any take home interviews. You should do the same. They don’t respect your time.
26.
by cochne 3y ago
No, GPT-4 is 1.7T params in the same way that an AMD Ryzen 9 7950X is (24 times X million transistors). It's not wrong to say GPT-4 is 1.7T parameters, your statement is wrong. Parameters can be configured in different ways, just as
27.
by cochne 3y ago
It’s not pedantry. Just because an inner product induced a metric space does it make it the same thing as that metric. The performance of similarity searches can depend very much on whether you use a dot product or a distance.
28.
by cochne 3y ago
Can you explain in what way Palantir is a criminal organization?
29.
by cochne 3y ago
The earth isn't a closed system. It receives energy from the sun which can be used to decrease entropy here on earth. Like an organism receiving energy from eating other organisms can regulate its own entropy. Sure you can't es
30.
by cochne 3y ago
Doesn't that imply a linear "n largest" algorithm? I guess that might already exist, but I wasn't aware of that!

More ›