evrydayhustling

6 ms·

1.
by evrydayhustling 13d ago
> weights update over time and past learning improves future learning such that we get better sample efficiency Is there an architecture-independent definition of forward transfer? For the practical experience and implications of AI prog
2.
by evrydayhustling 2mo ago
Around 2012 I was working near 5th Ave and 14th st, and there was a giant water main break. They had to dig about three stories down in a large intersection, and you could walk near the edge and look in. It was extraordinary, like somethin
3.
by evrydayhustling 2mo ago
Disagree, words drift but they still mean things. Nobody building accounting software in the 90s called it AI (or even an expert systems, barring the most inventive salescritters). Besides what kind of slippery slope argument is this? All
4.
by evrydayhustling 2mo ago
Horizon was accounting software - where is the connection to expert systems? Sounds like corporations can have bad motivations, and can make bad software - no AI required.
5.
by evrydayhustling 3mo ago
That paper shows hallucinations can't be eliminated, due to approximation error. But it is completely compatible with hallucination becoming less probable as scale reduces that approximation error.
6.
by evrydayhustling 3mo ago
> A Blu-ray disc, game cartridge, or printed book cannot be remotely erased, edited, or deactivated. It is a physical object you can own, resell, lend, archive, or play offline indefinitely. Isn't this untrue with surprising frequen
7.
by evrydayhustling 3mo ago
Feels like this could use a data driven way to normalize reference questions (topic and wording). This is missing a lot of right-initiated topics like immigration, gender roles, and role of religion in government.
8.
by evrydayhustling 3mo ago
The willingness to throw capital at AI is definitely doing some crazy things, but this article has some bad takes on the data. > [Ratio of per-token cost to subscription cost] means Anthropic is subsidizing their enterprise customers by
9.
by evrydayhustling 3mo ago
Understood; these are the failure modes of cheap, weak tests in a system designed for expensive, high signal tests. I get that it is causing a real pain when these are introduced willy-nilly -- so is there a way for the medical system to e
10.
by evrydayhustling 3mo ago
"We shouldn't collect information because we don't know how to fit it into our care playbook" might be rational for a single patient, but it's a policy that will lock you into your current playbook. Our medical indu
11.
by evrydayhustling 3mo ago
Brains are efficient, but civilized humans aren't. In the USA, adults consume at a rate of about 10kW -- only 1-2% of that being the human's metabolism, the rest being HVAC, electrical devices, etc. For comparison, a modern front
12.
by evrydayhustling 3mo ago
All major model providers offer prefix caching, which is this.
13.
by evrydayhustling 4mo ago
> data centers can be built basically anywhere this is especially true for AI use cases, where compute is hugely more important than latency / bandwidth > you have to provide a bonus pool that goes dollar-for-dollar for any buyba
14.
by evrydayhustling 4mo ago
Friends on ML/AI hiring committees at top tier university are seeing foreign profs turn away record offers. Same for applied math relevant to material science. I expect you are right at the most specialized end of the spectrum (and ce
15.
by evrydayhustling 4mo ago
Besides the brutal impact on those already invested in the American research community, this is one more nail in the coffin when it comes to competing for new talent. What researcher in their right mind would move their research and their
16.
by evrydayhustling 5mo ago
Besides deeply unpredictable factors (like signal transmission), most users of higher-level abstractions do so without certainty about how the translation will be executed. For example, one of the main selling points of C when I was growin
17.
by evrydayhustling 5mo ago
An LLM model itself -- that is, the weights and the mathematical functions linking them -- does not tell you exactly how to train from data, nor how to generate an output. Instead, it describes a function providing relative likelihood(outp
18.
by evrydayhustling 5mo ago
This is well said and good illustration of why optimality a fragile concept. High impact improvements often involve reframing the goal.
19.
by evrydayhustling 5mo ago
had the same question! looks like it's another project called Drawmode[1] from the same group... [1] https://github.com/teamchong/drawmode
20.
by evrydayhustling 6mo ago
If I understand this correctly, the pain is because you depreciated the assets in the first few years, offsetting other tax liabilities, and now you must pay those back even if exiting the properties at a wash? If so, it seems like the unfa
21.
by evrydayhustling 6mo ago
You are describing products that change the information flow used to solve a problem. There are probably big gains there, but it requires everyone to come for the journey at the same time - for example, you must tell the board "there
22.
by evrydayhustling 7mo ago
It's not even "whoa we have ethics", it's just "this is a bad look for us".
23.
by evrydayhustling 8mo ago
Thanks for confirming. I liked the clarity of outline but the AI-speak of the prose was really a slog.
24.
by evrydayhustling 8mo ago
I think many people would agree that the pursuit of that connection is valuable, even if it is never completed. Many of the same people (like me) would say that the biggest enemy of that pursuit is thinking you've finished the job. Tha
25.
by evrydayhustling 8mo ago
So do humans asked to answer tests. The appropriate thing is to compare to human performance at the same task. At most of these comprehension tasks, AI is already superhuman (in part because Gary picked scaled tasks that humans are surpris
26.
by evrydayhustling 9mo ago
Your intuition is exactly correct. An investor with tax to offset can essentially access the same future upside at a discount However, this discussion will be a perfect introduction to "finances at this level", where about 60% of
27.
by evrydayhustling 10mo ago
Best lines in this article. But it doesn't get to IMO a very important point: why can't these processes easily be structured? Here are some good reasons: - Your process interacts with an unstructured external world (physical rea
28.
by evrydayhustling 11mo ago
Ok, so you define violence to include advocating for a law that tramples someone rights. The next person says that advocating for laws is its own inalienable right, so you trampled him. And the whole semantic redefinition snake just eats
29.
by evrydayhustling 1y ago
This is so true. When you get DSA wrong, you end up needing insanely complex system designs to compensate -- and being great at Testing just can't keep up with the curse of dimensionality from having more moving parts.
30.
by evrydayhustling 1y ago
Those human imperfections likely decrease randomness - for example leaving cards that started adjacent more likely to remain adjacent han by strict chance.

More ›