phonebucket

9 ms·

31.
by phonebucket 4y ago
I'm pretty disappointed with how this has been handled. Mainly in two ways: - The scale of the breach (leaking of PII and encrypted vaults, that URLs are not encrypted) has only come almost 4 months after the initial investigation. Whe
32.
by phonebucket 4y ago
This answer feels generated by ChatGPT. No human would side with ChatGPT so strongly.
33.
by phonebucket 4y ago
Humans do engage with backwards reasoning. But they are also capable of checking that the constructed justification does then imply the conclusion. ChatGPT apparently is not doing this on the basis of these examples.
34.
by phonebucket 4y ago
I like this strategy, but I don't think it's necessarily clear cut: while taking the queen is forbidden, the opponent also has the opportunity of putting their queen en prise. So you end up in this scenario where both players are
35.
by phonebucket 4y ago
Pet peeve of mine: articles using stats like 96% accuracy. If the test set had 4% deep fakes, and 96% legitimate videos, a model which always predicts legitimate video would score 96% accuracy, even if it were useless. Stats like precision,
36.
by phonebucket 4y ago
> If the authorship is unearned the supervisor is just stuffing author lists which is unethical. Or the 2nd-3rd authorship is earned, but doesn't count enough as a scientific contribution to go into a thesis. > If the student is
37.
by phonebucket 4y ago
> If your PhD student publishes 8 papers with you and you only then notice or claim that they cannot graduate that is an enormous failure in supervision and mentorship in every imaginable case There are open questions nature of these pu
38.
by phonebucket 4y ago
> We don't use email much. I have 32500 unread emails. That many unreads makes it sound like everyone is using emails except you!
39.
by phonebucket 4y ago
> equity is significantly overvalued relative to fundamentals Which fundamentals? The $117.9 billion in revenue and $39.4 billion of net profit in 2021 [1]? The price to earnings ratio of 9.22, which is far lower than the NASDAQ average
40.
by phonebucket 4y ago
More important than matters of intelligence: Musk’s public unpopularity means nothing to his power, whereas it meant everything to Truss’s.
41.
by phonebucket 4y ago
> Scientists go down their own, squirrely rabbit holes. At the risk of taking the analogy too far: I'd wager that most people go down rabbit holes, but still enjoy general discussion on how to best dig said holes. Also, please do re
42.
by phonebucket 4y ago
I'd also love an HN-like community for scientists. The closest I know of is SciRate [1], where arXiv papers are posted. Users can then upvote and discuss them. I don't feel it has reached critical mass, so it's still of limit
43.
by phonebucket 4y ago
It's not machine learning that's required to make computers better than humans at chess. Non machine-learning chess engines engines based on alpha-beta pruning and expertly-crafted evaluation functions have been stronger than all
44.
by phonebucket 4y ago
Minor nitpick, but this is Scientific Reports, not Nature. It's run by the same publishers as the journal Nature, but is a significantly lower impact journal.
45.
The Annotated Diffusion Model(huggingface.co)
2 points by phonebucket 4y ago|0 comments
46.
by phonebucket 4y ago
> Read research publications and try to implement them. In my opinion, jump straight into this! Learn prerequisites as you need them. I found Goodfellow's book [1] to be helpful to learn some basics. But don't think you need to
47.
Real Not Complex: Free Math Resources(realnotcomplex.com)
8 points by phonebucket 4y ago|1 comments
48.
by phonebucket 4y ago
> to check if every single coupling in their code is correct? The author is not addressing the couplings in their code itself, but the code couplings between their code and all libraries in the Julia ecosystem, which is an altogether mor
49.
by phonebucket 4y ago
The quote provided explicitly suggests that tooling around reducing TTFX should also improve. It’s not suggesting improved CPU power as the sole mitigation.
50.
Matrices as Tensor Network Diagrams(math3ma.com)
2 points by phonebucket 4y ago|0 comments
51.
by phonebucket 4y ago
Can anyone recommend any arXiv/paper links on the subject for someone with reasonable prerequisties (e.g. neural ODEs, physics informed neural networks and message passing)? The number of references in the article is a bit of an overlo
52.
by phonebucket 4y ago
But what's the rationale behind this specific $1 limit? If something were to keep me entertained for several hours with no advertising, I'd certainly consider that worth $1. Going down the free route would likely end me drowning i
53.
Insect Power (2009)(airandspace.si.edu)
1 points by phonebucket 4y ago|0 comments
54.
by phonebucket 5y ago
This is a fun article! Alternative avenues that complement the approaches shown: - Padé Approximants ( https://en.wikipedia.org/wiki/Pad%C3%A9_approximant ) can be better than long Taylor Series for this kind of thing. -
55.
by phonebucket 5y ago
This does look like a useful tool. Especially since `jq | less` is a very common pattern for me. Apologies if this might take things a little off-topic. But I can recall seeing "written in Rust" in the titles of quite a few HN pos
56.
by phonebucket 5y ago
I also use vim for R for mostly the same reasons. However, you do indeed get autocompletion and many IDE amenities with the language server protocol. Naturally it’s not at the same level as RStudio. But one tool to play with any language is
57.
by phonebucket 5y ago
> All these languages that compile to SQL seem to forget one fundamental problem, when this stuff fails (queries becomes slow etc). We are left to debug auto-generated SQL, which is a complete nightmare.. My first inclination when hittin
58.
by phonebucket 5y ago
> But maybe less well at “dull” problems like good documentation and API design. I'm inclined to agree with you here. To my mind, this means that there's ample scope for developers from many backgrounds to make positive contrib
59.
NGS – Next Generation Shell(ngs-lang.org)
41 points by phonebucket 5y ago|29 comments
60.
by phonebucket 5y ago
> ie learn a bunch of esoteric commands (:Git blah) so that you don’t need to remember the other set of original commands I use fugitive, but haven't remembered these commands. I map what they do to shortcuts of my choosing e.g. \ga

More ›