Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
currymj
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
43 ms
·
571.
▲
by
currymj
7y ago
not screen space, storage space in the game. you might have to delete some in-game files.
572.
▲
by
currymj
7y ago
it's good to see the "real numbers are fake" crowd out in full force!
573.
▲
by
currymj
7y ago
historically, Japan officially used Western name order when communicating in Western languages. but this year it was switched, with requests going out to media organizations to update their style guides.
574.
▲
by
currymj
7y ago
in a world where you have many options and have to figure out which is best by repeated experimentation, but where experimentation itself has some cost, you have a multi-armed bandit problem. (the name is supposed to evoke a room full of sl
575.
▲
by
currymj
7y ago
in parts of American evangelical Christianity there’s a longstanding interpretation of the “mark of the beast” (which in the Book of Revelation is some sort of sign or mark without which people are forbidden from engaging in trade) as some
576.
▲
by
currymj
7y ago
it's very hard to buy right now, there's been a run on them due to coronavirus, and the rings are manufactured in China so supplies may not be replenished. They're going for hundreds of dollars.
577.
▲
by
currymj
7y ago
they put the biggest improvement upfront lol: things now scale properly if you have a 4K monitor. people (myself included) have been squinting at very tiny font for years, and now they don't have to anymore!
578.
▲
by
currymj
7y ago
the neural network stuff is very good and relies on GPU acceleration. it is the "computation graph" paradigm -- have to define the whole architecture up front -- but it's actually quite pleasant to use.
579.
▲
by
currymj
7y ago
it seems like this is the response from most people everywhere about until their region hits the knee of the exponential curve. this pattern played out in Italy, Spain, France, and now the US. the Imperial College paper estimates 2 million
580.
▲
by
currymj
7y ago
growth mindset: drinking is just a particular behavior, it's totally within my control, and all i have to do is learn how to make the choice not to drink irresponsibly. fixed mindset: i'm an alcoholic, i'll always be an alcoh
581.
▲
by
currymj
7y ago
maybe climbing was the better career move. if astrophysics is anything like most of academia, there are a lot of low-paying, unstable short-term positions as everybody scrambles for the few permanent jobs. i'm sure there are a lot of a
582.
▲
by
currymj
7y ago
i don't know to what extent the "best practices" even apply in most of these settings. pretty much everything about good, clean code is either not relevant or needs to get thrown out the window as soon as it hurts performance
583.
▲
by
currymj
7y ago
It seems to be a shibboleth specific to numerical computing people -- ODE solvers, linear algebra algorithms, or Fortran weather simulators tend to be described as "codes". Numerical applications are the biggest part of HPC and th
584.
▲
by
currymj
7y ago
it’s a scientific paper from a computer vision conference, it would be absurd in that context to assume anyone reading it doesn’t know that it stands for convolutional neural network. they didn’t write this with Hacker News in mind.
585.
▲
by
currymj
7y ago
the first two words are a hyperlink to a much longer paper.
586.
▲
by
currymj
7y ago
they add a regularization term to smooth things out and make the derivative exist. in a sense it’s similar to how argmax is a very discrete operation, but softmax is a good approximation that is differentiable.
587.
▲
by
currymj
7y ago
it’s not completely unprecedented because there were other ways of getting equivalent results before (the Sinkhorn based optimal transport approach cited, for one), which have been used for all kinds of interesting tasks. the contribution i
588.
▲
by
currymj
7y ago
schools at different levels of selectivity informally have some kind of SAT percentile cutoff but it's not ever a hard requirement, or the only thing considered, and also you can get into a reasonable university with a mediocre SAT sco
589.
▲
by
currymj
7y ago
it feels like many places have a property like this. i think it must be some combination of the way things get built around roads + transit development, and induced demand.
590.
▲
by
currymj
7y ago
at least in the US, income tax brackets are done in a mostly reasonable way. if the tax bracket is at $30,000 and you make $35,000, then you’d pay a low percentage of the first $30k and a higher percentage of the last $5k. so you can’t lose
591.
▲
by
currymj
7y ago
graph neural networks do generally seem to work pretty well for many tasks and are not snake oil. so it’s not unbelievable that they are also able to predict molecule properties with some degree of accuracy — and a fair number of papers hav
592.
▲
by
currymj
7y ago
this is increasingly common in (parts of) CS, see openreview.net. they even allow random anonymous comments from non-reviewers.
593.
▲
by
currymj
7y ago
they have to make sure you don’t work for the government or government contractors whose employees are prohibited from making political donations.
594.
▲
by
currymj
7y ago
for a lot of people who end up in research-type jobs, a sense of curiosity is one of their strongest motivators, and they want work that will let them pursue their curiosity. it sounds like you're motivated by something else.
595.
▲
by
currymj
7y ago
People can be cutthroat about trying to get their name on papers to which they contributed little, and they may be reluctant to discuss work too openly with others lest the discussion be construed as a collaboration which would require coau
596.
▲
by
currymj
7y ago
I think it would be very unethical for an undergraduate who contributed to a research effort not to be listed as an author. normally the professor's name goes last on the paper, and the convention is well understood that being last aut
597.
▲
by
currymj
7y ago
i’ve only run into stuff like that trying to get RL agents to train. Maybe GANs and other “hard to train” models are bad too idk. But generally things do actually seem to be reproducible.
598.
▲
by
currymj
7y ago
I would argue most machine learning papers that use public datasets have code available and are often also reproduced independently (sometimes just because of somebody's need to port between PyTorch/TensorFlow). Reproducibility is
599.
▲
by
currymj
7y ago
The states can legalize marijuana, meaning that the state police and courts no longer prosecute it. But currently the federal government can still come in and make arrests, although they typically choose not to enforce the law, because doin
600.
▲
by
currymj
7y ago
online learning as a term might actually predate any internet-based distance courses. Famous Freund and Schapire paper goes back at least to 1997 and uses the term in the title.
More ›