zmgsabst

16 ms·

301.
by zmgsabst 1y ago
As I pointed out, eg, the high number of correction terms when trying to tune the model to actual particle accelerator data is evidence that our model is missing something. (And some things are plain missing: neutrino behavior, dark matter,
302.
by zmgsabst 1y ago
That’s not my point, nor close to what I said. My point is that their models are fitted tensors/probability distributions, often retuned to fit new data (eg, the epicyclic nature of collider correction terms) — the same as fitting a DN
303.
by zmgsabst 1y ago
> This is not a failure of the experts to understand what's going on. > We don't have a model of why it works / if there's a more comprehensible layer of reality below it. Counterpoint: You’ve just admitted they do
304.
by zmgsabst 1y ago
US Dept of Ed has a $238B budget in 2024. I think much of that is at the collegiate level, but $53M is almost certainly wrong. https://en.wikipedia.org/wiki/United_States_Department_of_Ed...
305.
by zmgsabst 1y ago
I was curious, so I found random numbers on Google: - a Panmax ship is 8000 sq m (250m long x 32m wide) - using 150w per sq m this is 1.2Mw - a Panmax ship has a 40,000hp engine, which is 30Mw So you’re off by a factor of around 20x using s
306.
by zmgsabst 1y ago
I wish the formality would be included in an appendix — as someone who has had to implement a lot of things (and more than once, found errors). But I agree with your general point: understanding the recipe and general thrust of the approach
307.
by zmgsabst 1y ago
I’m not making any claim, but it’s clear the people accused of removing them feel the bias delegitimizes the results — so the OP’s logic is incorrect. As I originally said. Also, factually, they’re merely transferring the website to a diffe
308.
by zmgsabst 1y ago
I was just pointing out the incorrect logic of the comment: If you believe they’re biased, removing them is removing false legitimacy from biased content — something the government has an interest in. And no different than a journal retract
309.
by zmgsabst 1y ago
That’s what category theory does well: broad collections of trivialities in a unified definition.
310.
by zmgsabst 1y ago
US spent just under $2T in Iraq and just over $2T in Afghanistan, for a total of just over $4T.
311.
by zmgsabst 1y ago
I’ve found managing the context is most of the challenge: - creating the right context for parallel and recursive tasks; - removing some steps (eg, editing its previous response) to show only the corrected output; - showing it its own out
312.
by zmgsabst 1y ago
“Compete” is an economics term. Making it cost more is keeping them from being as competitive — which is the point of sanctions.
313.
by zmgsabst 1y ago
Are you saying most humans aren’t generally intelligent, by your definition?
314.
by zmgsabst 1y ago
I’d argue we have AGI, at the level of a child; now we’re debating further steps, such as adult AGI and super intelligence. But because AI is not like us, we have different results at different stages — eg, they’ve been better at arithmetic
315.
by zmgsabst 1y ago
You’d know the value in a more powerful system than ZFC (as it includes such an oracle) — but you can already reason about ZFC in a more powerful system. We already have more powerful systems, but what causes the inability to self-reason is
316.
by zmgsabst 1y ago
Americans see tribalism just fine: we’ve been discussing since Malcolm X how the tribal sentiments of minorities are utilized by a political party in the pursuit of power. And the primary division these days is urban vs rural, with the seco
317.
by zmgsabst 1y ago
I didn’t introduce it, you did — by positing that formal language is more objective, as you’ve again done here. My original point was that mathematics is human language. > The further point is that the relative complexities of two system
318.
by zmgsabst 1y ago
NSA is a DOD organization. > The National Security Agency (NSA) is an intelligence agency of the United States Department of Defense, under the authority of the director of national intelligence (DNI). https://en.wikipedia.org
319.
by zmgsabst 1y ago
The structure that most words describe is objective, so you haven’t distinguished math as a language. (Nor is mathematics entirely “objective”, eg, axiom of choice.) And the number of terms in your chosen language with your chosen basis isn
320.
by zmgsabst 1y ago
Mathematics is a human language. It being a formal language doesn’t change that. Further, it’s not objective: you’re choosing the basis which causes the complexity, but any particular structure can be made simple in some basis.
321.
by zmgsabst 1y ago
“Number of terms” is a human language construct.
322.
by zmgsabst 1y ago
My experience is the opposite: I haven’t worked somewhere we ran code locally in a long, long time. Your IDE is local, but the testing is remote — typically in an environment where you can match the runtime environment more closely (eg, ens
323.
by zmgsabst 1y ago
Sometimes it’s nice to frame it the other way, eg: 90% -> 1 error per 10 99% -> 1 error per 100 99.99% -> 1 error per 10,000 That can help to see the growth in accuracy, when the numbers start getting small (and why clocks are fram
324.
by zmgsabst 1y ago
Developers are generally given specific environments to run code, which aren’t their laptops — eg, VMs in a development environment. The goal isn’t to stop a developer from doing something malicious, but to add a step to the chain for hacke
325.
by zmgsabst 1y ago
Sure — but now we’re back in the case the person above me was calling out and I was emphasizing by pointing out the stochastic nature: you moved the goal posts. From their comment: > Which is why it's not really incredible that it c
326.
by zmgsabst 1y ago
All infinite sequences are physically impossible, but they’re the basis of asymptotics; arbitrary failure and non-production of entropy are still physically possible.
327.
by zmgsabst 1y ago
You have a source of randomness — you just got exceptionally unlucky. A truly random source can produce the same value indefinitely. My point is that you have stochastic termination, but not guaranteed termination. Those are different thing
328.
by zmgsabst 1y ago
Yes you are — you just have exceptionally bad luck to generate that sequence.
329.
by zmgsabst 1y ago
The difference is deterministic termination: There is no N such that your algorithm is guaranteed to terminate before N flips — even for a single bit. The complexity class in the worst case (100% T) is infinite; it’s only the average cases
330.
by zmgsabst 1y ago
Citing a few points to justify my own conclusion: > Many (not all) humans screw up on versions of the Tower of Hanoi with 8 discs. > LLMs are no substitute for good well-specified conventional algorithms. > will continue have their

More ›