jampekka

42 ms·

931.
by jampekka 2y ago
He did get the Turing award already in 2018.
932.
by jampekka 2y ago
Hinton did introduce the Boltzmann machine networks that kind of resemble thermodynamics if you squint really hard. And those didn't really pan out. Of course the price is really for MLP backpropagation (or how it found applications) b
933.
by jampekka 2y ago
Except if/when it's cached.
934.
by jampekka 2y ago
Security through complexity.
935.
by jampekka 2y ago
Primitive-based representations are a lot easier to manipulate (e.g. animate) than NeRFs. Also they can be a lot more efficient/scalable when there's a lot of e.g. empty space.
936.
by jampekka 2y ago
What is robbery is defined in law, and by law taxation is not robbery.
937.
by jampekka 2y ago
It is just-so. That is the primary method in evolutionary psychology. https://en.m.wikipedia.org/wiki/Just-so_story
938.
by jampekka 2y ago
OpenAI LLMs are some of the least open.
939.
by jampekka 2y ago
At least for now people haven't at large elected politicians who would eliminate taxes.
940.
by jampekka 2y ago
Those civil servants execute laws that are, at least ideally, democratically decided upon. In contrast to private concentrations of wealth making arbitrary decisions. And many of those laws are forcibly denying people from access to somethi
941.
by jampekka 2y ago
JSTOR is a disgrace. RIP Aaron Swartz.
942.
by jampekka 2y ago
After pointless hoops. And the process seems to involve pointless hoops even when the review is not rejected. https://github.com/uBlockOrigin/uBOL-home/issues/197
943.
by jampekka 2y ago
Maybe a less crappy review system at least? "The burden is that even as a self-hosted extension, it fails to pass review at submission time, which leads to having to wait an arbitrary amount of time (time is an important factor when al
944.
by jampekka 2y ago
I can see how having to jump pointless bureaucratic hoops in a volunteer project can cause throwing out toys.
945.
by jampekka 2y ago
Social safety net causes deadweight loss. To be clear: I'm not for minimizing deadweight loss because free markets are horrible for humans. Free markets don't care about human suffering. Regulated markets can be OK, but then you s
946.
by jampekka 2y ago
Or just pirate new games while you're at it? (Which I fully support.)
947.
by jampekka 2y ago
Banning slavery caused dead-weight loss. Market clearing price for commodified labor is minimal sustinence.
948.
by jampekka 2y ago
If a hotel cleaner steals your laptop, the hotel should not be responsible?
949.
by jampekka 2y ago
Austrian economics also rejects empirical assesment of its claims. Instead, universal thruths are derived "logically" (formal logic banned though) from "obviously true" axioms using a method called praxeology. It seems a
950.
by jampekka 2y ago
> China folks are more hungry than most in the west This is a big part, but China also has a different economic model where they can plan ahead and invest for long term scenarios and strategies. Shareholder value maximization maybe wasn&
951.
by jampekka 2y ago
Western media keeps on being surprised over and over that Chinese can compete with and surpass western technology. In manufacturing they've been ahead for decades, and it's nowadays very common to find state-of-the-art software fr
952.
by jampekka 2y ago
Occam's razor says they are saying whatever gives them as much money and power as possible. OpenAI is suspiciously bad in providing access to their models, compared to e.g. Meta, Mistral, BLOOM collaboration, Alibaba and even oil tyran
953.
by jampekka 2y ago
OTOH humans can e.g. walk on two feet and drive a car.
954.
by jampekka 2y ago
No dev tools are readily available on Windows.
955.
by jampekka 2y ago
A prime example how all the paranoid security hoops can easily make things more insecure in aggegate. Since Microsoft embracing and extending it, GitHub has become one of the worst offenders.
956.
by jampekka 2y ago
The paper itself doesn't claim it's AI. They do say it's "machine learning". https://www.cmaj.ca/content/196/30/E1027
957.
by jampekka 2y ago
This method has nothing to do with LLMs or even "deep learning" though.
958.
by jampekka 2y ago
It is based on a relatively traditional time series regression method. "AI" is just the usual spin.
959.
by jampekka 2y ago
GOFAI from 1991! Wasn't AI back then though. https://en.m.wikipedia.org/wiki/Multivariate_adaptive_regres...
960.
by jampekka 2y ago
E.g. OpenGL normalized device coordinates have x=0,y=0 at screen center, with top left being x=-1,y=1. It's easier to reason with in many cases.

More ›