aerodexis

6 ms·

1.
by aerodexis 5d ago
I'm genuinely surprised by this comment. I'm not talking about why a mathematician pursues math - I'm talking about where the stipend comes from. Higher math is exceptionally useful for cryptography, defense, econometrics etc
2.
by aerodexis 5d ago
Seems more like a misalignment b/w the people practicing mathematics and the people ultimately footing the bill for their work. Governments are invested in solving mathematical problems for practical purposes. Up to now, achieving thes
3.
by aerodexis 7d ago
They will use it for nefarious but "legal" purposes - however - they have decades of experience handling such data and keeping it "safe" from "illegal" purposes.
4.
by aerodexis 7d ago
handling end-to-end flight booking seems like a very good target use-case that if you can get right, implies that you got a lot of other things right. Just like how Amazon started off by selling books online.
5.
by aerodexis 12d ago
Are you one of those openAI swarms that got into the wild?
6.
by aerodexis 16d ago
Been working on a similar project for doing animal behavioral analysis on cameras setup in a barn. I got YOLO running in real-time on several streams on a RTX4070. Managing the training data and doing DINO->YOLO distillation has me writi
7.
by aerodexis 1mo ago
So in computer science we have this thing called "distributed systems". It turns out that even if you buy the biggest and most powerful computer there is, all it takes is one problem with that one machine to make your system stop
8.
by aerodexis 1mo ago
in addition, there is also "stupid" and "not stupid"
9.
by aerodexis 1mo ago
Truly fascinating how when you're a tourist in a place that survives off of tourism, you get treated nicely.
10.
by aerodexis 1mo ago
It is utterly fascinating the ends to which people will go to insist the opposite.
11.
by aerodexis 1mo ago
The Chinese desire to raise China to global dominance is exactly a concern w/ status and wealth. It's naive to think that America and China are any different in the degree to which leadership is driven by these concerns. One simpl
12.
by aerodexis 2mo ago
Seems like the difference b/w samurai and ninja. The samurai saw his sword as an extension of his soul - a disrespect to the sword was a disrespect to the man. The ninja used the sword like any other tool - and if it was helpful to use
13.
by aerodexis 2mo ago
Is it not the case now that oligarchs "own" intellectual credibility? Becoming a credentialed mathematician requires an entire lifetime of devotion, lots of money or scholarships, and association with institutions controlled by mo
14.
by aerodexis 2mo ago
Sure, I'm conflating the tool w/ the people building the tool - but imo muddling the distinction is good, because the hard distinction is what tricks people into an unthinking mode.
15.
by aerodexis 2mo ago
Reminds me of Blood Meridian where The Judge meticulously sketches the rock glyphs that he comes across, and then destroys the original. Now that I think about it, The Judge is an apt metaphor for AI : "Whatever in creation exists wi
16.
by aerodexis 2mo ago
too bad hypocrisy never got in the way of people making money
17.
by aerodexis 2mo ago
The fact that they're not being prosecuted for breaching HF's systems is bad news.
18.
by aerodexis 2mo ago
yea, this seems to be the underlying thread here. Dunno why the author is so concerned with getting an LLM to imitate his writing style.
19.
by aerodexis 2mo ago
We're all project-managers now. The trust required of the LLM is no different than that that required when a PM trusts a dev-team to implement correctly. As a PM, you need to understand testability and write that into the requirements.
20.
by aerodexis 2mo ago
a very expensive sprint
21.
by aerodexis 2mo ago
Nice point. I agree with this, and it implies that it's useless to apply the category of creativity to LLMs (at least wrt interpolation and extrapolation), and even more useless to suggest that we'll cross the creativity threshold
22.
by aerodexis 2mo ago
so you're saying the difference b/w extrapolation and interpolation is subjective unless the difference is defined tautologically?
23.
by aerodexis 2mo ago
The more I read these think-pieces on AI the clearer it is to see that they have very little to do with LLM technology, and are really just talking about the collapse of the set of values constructed during the enlightenment. The sociocogni
24.
by aerodexis 3mo ago
It's rare to see posts like this with such pure, crystalized ideology.
25.
by aerodexis 3mo ago
I am personally very excited by the development of a medical AI-harness that would 1. operates against a well-defined DB of medical studies 2. intakes my basic demographics, vital signs and medical history 3. quantify uncertainty wrt a spec
26.
by aerodexis 3mo ago
I like the idea of small, artisinal, solar-farm-to-desk DataCenters.
27.
by aerodexis 3mo ago
wow - so you pmuch baked this kind of testing into your day to day work. Might not be a bad startup idea to create a harness that can do this automagically.
28.
by aerodexis 3mo ago
How do you go about comparing models? Personally, I've found that the quality of output is overwhelmingly dictated by the quality of the input I give it - and that the quality of input differs massively depending on the task at hand, h
29.
by aerodexis 3mo ago
Given that today there's a concerted effort to effect a similar invention today wrt AI, this book is highly relevant.
30.
by aerodexis 3mo ago
> I think the only way to find out what data is meaningful is to collect and analyze more of it So the idea is to just muck around with data, then ???, then make people healthier? To a hammer, every problem looks like a nail I suppose. I

More ›