auntienomen

9 ms·

61.
by auntienomen 2y ago
The difference is that the subway is the main transit mode. OP is very reasonably complaining that every reasonable public transit option to NYC airports is multi agency and multi modal.
62.
by auntienomen 2y ago
One of my favorite things about the statistical physics of water is that H2O is a strong electric dipole. You'd expect this plays an important role on its nucleation, and yet we get quite good approximations when we neglect this fact.
63.
by auntienomen 2y ago
All true, but using a BigDecimal to do that is abusing the implementation and keeping important assumptions implicit. Better to explicitly quantify the uncertainty with a second variable.
64.
by auntienomen 2y ago
This seems like incredibly bad design. Is there any justification or use case where someone would want the abstraction to leak like this?
65.
by auntienomen 2y ago
A good DS can double as an MLE.
66.
by auntienomen 2y ago
Time will tell.
67.
by auntienomen 2y ago
Yeah, category theory is a framework for describing mathematical structure. It's not vacuous -- there are mathematical structures which don't fit into the framework and there are some theorems about what properties a category ha
68.
by auntienomen 2y ago
This is a 5 paper long argument aimed at other mathematicians. It's a formal proof in the sense that it's standard math talk, but not a formally verifiable proof that could be, e.g., machine checked.
69.
by auntienomen 2y ago
EGO makes a really good one.
70.
by auntienomen 2y ago
The money doesn't belong to the heirs. Winner can donate all of it if he wants.
71.
by auntienomen 3y ago
Footnote: Haag's theorem only invalidates naive perturbative approximation. But that's not what physicists actually use; they use it for motivation and then switch to a renormalized perturbative approximation, which is compatibl
72.
by auntienomen 3y ago
The fine seems reasonable. He disgorges the profits, and then gets fined twice what he earned and banned from serving on boards (worth far more than $1m over 5 years). Big question for me is why was he doing crimes for such low stakes.
73.
by auntienomen 3y ago
It's not necessary to be in favor of anything, in order to criticize what's already out there. You can point out fatal flaws in an idea without needing to have an alternative in mind. Woit himself is working (pretty casually to m
74.
by auntienomen 3y ago
It's the best place I know of for math gossip.
75.
by auntienomen 3y ago
This is quite nice. Any plans to add simpler ML tools? A sklearn plugin could be as valueable as LLM access.
76.
by auntienomen 3y ago
They're saying companies pay the costs of RTO in order to have more control over their employees.
77.
by auntienomen 3y ago
RTO is just backdoor age and parental status discrimination. Gotta shake out everyone who's not a cheap twentysomething.
78.
by auntienomen 3y ago
Sleep apnea is extremely common. And it appears to be correlated with having had corrective orthodontics, so may in fact be more common than previously.
79.
by auntienomen 3y ago
A bit older than that. This joke predates Bell labs.
80.
by auntienomen 3y ago
"Options/RSUs with actual value" is a very nice way to put it.
81.
by auntienomen 3y ago
I suppose a log-finitist takes O(1/n) to answer for 2*n? :-)
82.
by auntienomen 3y ago
Drunk driving is worse than stupid. It's up there with shooting a gun into someone's house.
83.
by auntienomen 3y ago
Winter holiday celebrations & festivals predate the major world religions. Don't presume there's no meaning for people who don't follow a particular religion.
84.
by auntienomen 3y ago
I found Cosma Shalizi's notes on the subject pretty insightful. http://bactra.org/notebooks/nn-attention-and-transformers.ht... Definitely read through to the last section.
85.
by auntienomen 3y ago
It's a winning hand, but your odds of drawing it are inverse galactic.
86.
by auntienomen 3y ago
What would be nice is an LLM to aid in translation between mathematics and a theorem proving language like Lean. That might fuel a revolution in how mathematics is done.
87.
by auntienomen 3y ago
Open conjectures are usually solved by inventing new mathematical frameworks that embed the context of the existing question. Inventing new definitions is trivial, but there's no rules that constrain the space of moves. And inventing
88.
by auntienomen 3y ago
Sample size 1. Here's what Andrew Wiles, the only other person to have solved a Millennium-class problem has to say of math competition: "Let me stress that creating new mathematics is a quite different occupation from solving pro
89.
by auntienomen 3y ago
Yeah, but none of that gets you a solution to a Millennium problem. One needs an AI to do something like what Peter Scholze did in inventing perfectoid spaces or Perelman in introducing his entropy function. You can treat axiom invention
90.
by auntienomen 3y ago
We can. But doing so frames math research as the same sort of activity as math problem solving.. it's not. Many imo champions struggle to do any successful math research. And many successful math researchers (e.g., all of the most

More ›