jmoss20

9 ms·

31.
by jmoss20 4y ago
? What contribution are you imagining that involves a PR to all 2500 projects?
32.
by jmoss20 5y ago
org mode?
33.
by jmoss20 5y ago
Well, you also have to be /correct/. There's some tension between correctness and clarity. Both come in degrees, and writing in the top right quadrant is rare. But holding correctness fixed, I think OP (and Feynman) are right
34.
by jmoss20 5y ago
I don't agree with the article, but hold on -- you don't need to buy "intelligence -> soul" for this. There are plenty of things that can't (or can't efficiently) be represented in a single agent. This is (f
35.
by jmoss20 5y ago
But brains do, in fact, compute things! There's no reason to think that they compute things the same way silicon "computers" do -- that they're arranged in a von Neumann architecture, or something. But it is true that th
36.
by jmoss20 5y ago
> But that fact isnt sufficient to make "semantic distinctions" -- because the proposition isn't true in virtue of that frequency. The trick here is that language models are (currently!) demonstrating you /can/ g
37.
by jmoss20 5y ago
> the college wage premium is fatter than ever, The more accessible (and popular) college is, the bigger signal it is that someone didn't go. That premium could just as well be the result of the (relative) wage floor falling out for
38.
by jmoss20 5y ago
It's easier to build solar farms than solar cars. (wind farms than wind powered cars, nuclear plants than nuclear powered cars...)
39.
by jmoss20 5y ago
Relative to their peers though: Alphabet went up almost 8% the same day, clearly not due to Microsoft's VR contract.
40.
by jmoss20 6y ago
As far as I understand, the first claim: > [capitalist states] consciously constrains the number of available jobs, maintaining a stable unemployment % among the populace. is true, and not particularly controversial. Too little unemploym
41.
by jmoss20 6y ago
What is transfer learning if not partially updating the model...?
42.
by jmoss20 6y ago
Unclear which way the causal arrow goes.
43.
by jmoss20 6y ago
Can you say more? Is there a thermodynamic interpretation of, say, the Halting problem? (Or P vs. NP?)
44.
by jmoss20 6y ago
> Eventually we'll have an a-ha moment... Will we? How can we ever verify that our intervention produces this or that sort of qualia? (For the record, I'm a fan of theories like IIT [0] that aim to answer questions in the neig
45.
by jmoss20 6y ago
Sure, but the parent's claim was that there is no "want" and "like" in evolution . You might want and like things; perhaps ( perhaps ) cells and bacteria might want and like things, but evolution certainly does not
46.
by jmoss20 6y ago
At least one reason: universities are overwhelmingly funded by public money. Perhaps (/perhaps/) private institutions should be allowed to explicitly advantage wealthy applicants, but forcing all of us fund institutions that advan
47.
by jmoss20 6y ago
Read the article. 64 students (from the tiny audited sample) met the highest bar (direct email evidence of connection to donations etc.), to say nothing of other "strongly suspect" students, and the rest outside of the audited sam
48.
by jmoss20 6y ago
That's exactly what the study implies. (It doesn't try to correct for those variables; it doesn't have to.)
49.
by jmoss20 6y ago
> The president also theoretically has the power to just unilaterally start a war... ? " The Congress shall have Power . . . To declare War, grant Letters of Marque and Reprisal, and make Rules conquering Captures on Land and Water
50.
by jmoss20 6y ago
(Air Force brat): There's a lot of interesting game-theoretic stuff here too. If none of your prisoners dilemmas are iterated (or only iterated few times), defecting is always advantageous. There are lots of tradeoffs with "long
51.
by jmoss20 6y ago
Sure, but Harvard's position in society is no more fixed than anyone else's (albeit with more inertia). They maintain their status by producing high status graduates. If we want to influence that dynamic (say, to push it toward ju
52.
by jmoss20 6y ago
1) The incentives are different for different universities. Less elite universities don't (generally) do all this because, you're right, there's no point. More elite universities do, because they're incentivized to. El
53.
by jmoss20 6y ago
Credentials vs. education. Credentials are a positional good, education an intrinsic one. We've decided the two should be bundled together, but they don't have to be. Maybe they shouldn't be; that would be a much more inte
54.
by jmoss20 6y ago
> X is the worst! > No, look at how bad Y and Z are! They're so bad! this is a... Sokal? (I think it's pretty easy to read the intent here, and more generally: internet comments are hardly rigorous proofs, and that's
55.
by jmoss20 6y ago
Doesn't this assume that morality has mostly to do with consequences (effects/outcomes/...)? That's a reasonable idea, but it's relatively modern. Semitic religions seem more concerned with (or at least, also very
56.
by jmoss20 6y ago
How would you update on it? (But you could just as well s/constant speed/average speed/, I don't think it makes a difference.)
57.
by jmoss20 6y ago
The upshot here is that choosing the max entropy distribution as your prior isn't enough, you also need to choose some particular way to formulate the problem. Particular formulations (travel time vs. speed, here) imply different max
58.
by jmoss20 6y ago
Thought experiment: suppose your friend drives 80 miles to visit you. They tell you the trip took between 2 and 4 hours. You have no further information. How confident are you the trip took less than 3 hours? Now they tell you they maint
59.
by jmoss20 6y ago
big Nozick guy huh?
60.
by jmoss20 6y ago
You still haven't resolved the inconsistency.

More ›