Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)B

barsoap

@ barsoap @lemm.ee

Posts
9
Comments
743
Joined
3 yr. ago

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • To the point that you say things like the universe already is a formal system to which we can apply the incompleteness theorem.

    And that is contentious, why? If the laws of the universe are formalisable, then the universe is isomorphic to that formalisation and as such also a formal system. We're not talking being and immanence, here, we're talking transcendent properties.

    When I asked this, you only mentioned being able to express natural numbers. But can the formal system express them in the specific sense that we need here to use incompleteness?

    How do you express them in ways that do not trigger incompleteness? Hint: You can't. It's a sufficient condition, there's equivalent ones, if I'm not mistaken an infinite number of them, but that doesn't matter because they're all equivalent.

    These are all giant holes you skipped

    These are all things you would understand if I didn't have to remind you of basic computability and complexity theory literally every time you reply. As said: Stop the philosophising. The maths are way more watertight than you think. We're in "God can't make a triangle with four sides" territory, here, just that computability is a wee bit less intuitive than triangles.

    If you want to attack my line of reasoning you could go for solipsism, you could come up with something theological ("god chooses to hide certain aspects of the universe from machines" or whatever). I'm aware of the limits. I didn't come up with this stuff yesterday and my position isn't out of the ordinary, either.

  • China’s new bullet train prototype sets global speed record at 450 km/h in test run - Hindustan Times

    Jump

  • ...production train in revenue service over the period of usual track maintenance. It's not the rails I'm worried about but the overhead lines, have a look at the TGV record run there's a lot of arcing that can't be good for them. On the pantograph side the contacts are carbon, notoriously brittle, and generally speaking over 300km/h things become dicey just because of heat buildup. Automated retraction systems can prevent the worst, meaning the pantograph getting tangled in the lines at speed, but, well: Production revenue service. Can't have those fail-safes kick in or you'll mess up the timetable.

    European high speed trains actually got slower vmax over the years for the simple reason that the rest of the systems is reaching practical limits and it doesn't make sense to have trains that can drive speeds that the track will never reasonably support. If you want to go faster you want to go contactless which means building a whole new, incompatible, network. And the only maglev system with affordable tracks, as in bucks per kilometre (TSB), has a design limit of ~200km/h: It uses bog-standard third-rail like technology for electrification, everything else would be expensive. But for its intended use-case as an S-Bahn 200km/h are plenty and you get nice bonuses such as 10% gradients and inexpensive viaducts (because no point-loads they can be built way lighter).

    China, in the end, isn't that large. You can do high-speed sleeper trains, 12-14h or such, from one end of the country to another with existing stock technology, and so btw could the US.

  • Critics decry Vietnam’s ‘draconian’ new internet law

    Jump

  • Dictatorship doesn't imply that there's only a single ruler.

    That said the situation, as in the power of the Politburo, isn't as extreme as it was in the USSR or in Vietnam before 1988. But they still have ways to go before they're at Cuban levels of "wait we'll have to take a closer look they might actually have come up with a form of democracy constitutionally different from the usual ones". Cuba is still authoritarian but that seems to be more cultural inertia than tankie ideology. Singapore might actually be a good comparison, not economically but politically.

  • Critics decry Vietnam’s ‘draconian’ new internet law

    Jump

  • Technically correct, which is the most tone-deaf kind of correct.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • youtube expositions

    Dude any uni you go to likely has lectures that are worse than the Arsdigita ones. If you want to save face, act less like a philosopher next time and don't assume that you know things better than someone who actually studied it. I ended up linking veritasium because I realised that you have no idea about the mathematical fundamentals or you wouldn't say silly things such as

    entails that there are uncomputable functions, which I take to mean that there are uncomputable oracles in the physical world.

  • EU law mandating universal chargers for devices comes into force

    Jump

  • There's always an implementation period with these things, also with the USB thing, to allow companies to build and sell phones that are already in the pipeline. Expect, just as with the USB thing, replaceable batteries to become a common sight quite soon and ubiquitous by 2027. You can already get quite decent smartphones with replaceable batteries but it's the usual suspects Fairphone, Gigaset, and (at least one model of) Samsung, those would also exist without the regulation. The "oh shit they actually passed it we'll need to re-engineer things" models from everyone else still aren't on the market.

    And before anyone brings it up: Yes, you can make them waterproof.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • entails that there are uncomputable functions, which I take to mean that there are uncomputable oracles in the physical world.

    It means that there are functions that are not computable. You cannot, for example, write a program that decides, in finite time, whether an arbitrary program halts on a particular input. If you doubt that, have an easy-going explanation of the proof by diagonalisation.

    We take the set of laws governing the universe and turn it into a formal system. How?

    Ask a physicist, that's their department not mine. Also I'd argue that the universe itself is a formal system, and lots of physicists would agree they're onto the whole computability and complexity theory train. They may or may not agree to the claim that computer science is more fundamental than physics, we're still working on that one.

    Does the resulting formal system really meet all conditions of the incompleteness theorem?

    Easily, because it will have to express the natural numbers. Have a Veritasium video on the whole thing. The two results (completeness and in computability) are fundamentally linked.

    he turing machine wouldnt be adequate to model the brain in the sense that the brain, in that hypothetical scenario, would be a hypercomputer,

    If the brain is a hypercomputer then, as already said, you're not talking physics any more, you're in the realms of ex falso quodlibet.

    Hypercomputers are just as impossible as a village barber who shaves everyone in the village who does not shave themselves: If the barber shaves himself, then he doesn't shave himself. If he shaves himself, then he doesn't shave himself. Try to imagine a universe in which that's not a paradox, that's the kind of universe you're claiming we're living in when you're claiming that hypercomputers exist.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • You say an incompleteness theorem implies that brains are computable?

    No, I'm saying that incompleteness implies that either cause and effect does not exist, or there exist incomputable functions. That follows from considering the universe, or its collection of laws, as a logical system, which are all bound by the incompleteness theorem once they reach a certain expressivity.

    All I said is that the plain old Turing machine wouldn’t be the adequate model for human cognitive capacity in this scenario.

    Adequate in which sense? Architecturally, of course not, and neither would be lambda calculus or other common models. I'm not talking about specific abstract machines, though, but Turing-completeness, that is, the property of the set of all abstract machines that are as computationally powerful as Turing machines, and can all simulate each other. Those are a dime a gazillion.

    Or, see it this way: Imagine a perfect, virtual representation of a human brain stored on an ordinary computer. That computer is powerful enough to simulate all physical laws relevant to the functioning of a human brain... it might take a million years to simulate a second of brain time, but so be it. Such a system would be AGI (for ethically dubious values of "artificial"). That is why I say the "whether" is not the question: We know it is possible. We've in fact done it for simpler organisms. The question is how to do it with reasonable efficiency, and that requires an understanding of how the brain does the computations it does so we can mold it directly into silicon instead of going via several steps of one machine simulating another machine, each time incurring simulation overhead from architectural mismatch.

  • Insiders say cutting-edge graphics are too costly for AAA games

    Jump

  • The latest game where I thought "damn this looks good" was Sifu. I get like 200fps on my half-potato (5500XT), and that's at ultra quality, definitely "let's turn on vsync to get rid of the fan noise" territory. The reason it looks good is good lightening choices, fluid animation, as well as well-decorated levels. As you can see the textures and geometry are often very simple -- a red fire hose box in a a hallway is just a red box. No fine detail at all, and that's sufficient: It's enough detail so that things don't feel empty, your brain isn't thinking "there should be more here", a whole uncanny valley of its own as the brain gets kinda queasy if there's nothing that it can ignore, but not enough detail as to be cluttering, that is, detract from the readability of the graphics.

    Good style and execution will always win out over realism.

    And yes it's a 30G game, high-res textures and not kitbashing the levels tends to do that. Also, storing stuff uncompressed the download size is 20G.

    (And btw whoever made that video is a good player deliberately playing like ass. You can tell by how they're taking ages to get through the level, the pitiful score, but still not dying or really taking much damage at all).

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • Who is to say the brain doesnt take advantage of some weird physical phenomenon that isnt classically computable?

    Logic, from which follows the incompleteness theorem, reified in material reality as cause and effect. Instead of completeness you could throw out soundness (that is, throw out cause and effect) but now the physicists are after you because you made them fend off even more Boltzmann brains. There is theory on hypercomputation but all it really boils down to is "if incomputable inputs are allowed, then we can compute the incomputable". It should be called reasoning modulo oracles.

    Or, put bluntly: Claiming that brains are legit hypercomputers amounts to saying that humanity is supernatural, as in aphysical. Even if that were the case, what would hinder an AI from harnessing the same supernatural phenomenon? The gods?

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • But we know too little about whether the limits of the turing machine are also limits of human cognition.

    Erm, no. Humans can manually step interpreters of Turing-complete languages so we're TC ourselves. There is no more powerful class of computation, we can compute any computable function and our silicon computers can do it as well (given infinite time and scratch space yadayada theoretical wibbles)

    The question isn't "whether", the answer to that is "yes of course", the question is first and foremost "what" and then "how", as in "is it fast and efficient enough".

  • Even Apple wasn’t able to make VR headsets mainstream in 2024

    Jump

  • (besides perhaps valve)

    Definitely besides valve, they didn't lose their mojo as gamedevs. In fact valve is what happens when gamedevs have too much money: Too few fucking games.

  • The Paper Passport Is Dying

    Jump

  • You probably can, not just in non-EU Schengen countries (i.e. EFTA) but also a couple of other select places. E.g. Egypt will let you in with a German ID card if you have two extra images with you so they can issue you a small cardboard visa. Wouldn't recommend it in Egypt though, banks, hotels etc. might not recognise ID cards.

    But that's really the main issue: The country will have to issue a visa and that has to be recorded in some way. It could, in principle, be completely electronic and online, but that requires that their IT systems can actually use the electronic features of your ID card and that everyone who might have to check your visa has to have a card reader and a connection to the state's servers.

  • AI Needs So Much Power, It’s Making Yours Worse

    Jump

  • Renewables won't exactly help harmonics, on the contrary, especially solar. This is an issue of insufficient mitigation mechanisms, probably on the supply side as computer PSUs are generally quite well-behaved loads: Drawing lots of electricity, on its own, does not harmonic distortion make.

    If it is on the consumer side utilities need to start charging commercial customers for distortions just like they're charging for blind current. If it's on the supply side, utilities need to require large solar installations to have proper filters, and have their own mechanisms to mop up the rest. Generally the US should start having a not shoddy electricity grid, brown- and blackouts and you call yourself a developed country? We don't even have a (colloquial) word for brownout over here!

    That all said, yeah the AI hype gotta stop. That doesn't mean that you should blame them for everything.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • One of the most successful application of LLMs might actually be quite enlightening in that respect: Language translation. B2 level seems to be little issue for LLMs, large cracks can be seen in C1, and forget everything about C2: Things that require cultural context. Another area where they break down is spotting the need to reformulate, that's actually B-level skills. Source: Open a random page on deepl.com that's not in English.

    Like, this:

    Durch weniger Zeitaufwand beim Übersetzen und Lektorieren können Wissensarbeitende ihre Produktivität steigern, sodass sich Teams besser auf andere wichtige Aufgaben konzentrieren können.

    "because less time required" cannot be a cause in idiomatic German, you'd say "by faster translating". "knowledge workers"... why are we doing job descriptions, on top of that an abstract category? Someone is a translator when they translate things, not when that's their job description. How about plain and simple "employees" or "workers". Then, "knowledge workers can increase their productivity"? That's an S-tier Americanism, why should knowledge workers care? Why bring people into it in the first place? In German thought work becoming easier is the sales pitch, not how much more employees can self-identify as a well-lubricated cog. "so that teams can better focus on other important tasks"? Why only teams? "improvements don't apply if you're working on your own?" The fuck have teams to do with anything you're saying, American PR guy who wrote this?

    ...I'll believe that deepl understands stuff once I can't tell, at a fucking glance, that the original was written in English, in particular, US English.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • Though I still think that an intelligent consciousness could emerge from a loop of generative “thoughts”, the most important of those probably being language.

    Does a dog have the Buddha nature?

    ...meaning to say: Just because you happen to have the habit of identifying your consciousness with language (that's TBH where the "stuck in your head" thing came from) doesn't mean that language is necessary, or even a component of, consciousness, instead of merely an object of consciousness. And neither is consciousness necessary to do many things, e.g. I'm perfectly able to stop at a pedestrian light while lost in thought.

    I don’t think that the phrase is “I think therefore I am”, but “I can think ‘I think therefore I am’ therefore I am”.

    What Descartes actually was getting at is "I can't doubt that I doubt, therefore, at least my doubt exists". He had a bit of an existential crisis. Unsolicited Advice has a video about it.

  • I never knew that USB-C extensions are not allowed for a reason

    Jump

  • charging electronics and just directly forward the voltage coming from the charger to the device

    I am highly sceptical of anything that would connect USB voltage, no matter how finely negotiated, directly to the battery terminals. Finely tunable voltage over USB is useful for keeping the buck/boost converter on the device side small, though, or just efficient because it doesn't have to do as much work. If you can charge over standard PD extending to charging over PPS should only be a software change as your hardware is already more than capable.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • It's a basic argument of generative complexity, I found the article some years ago while trying to find an earlier one (I don't think by the same author) that argued along the same complexity lines, essentially saying that if we worked like AI folks think we do we'd need to be so and so much trillion parameters and our brains would be the size of planets. That article talked about the need for context switching in generating (we don't have access to our cooking skills while playing sportsball), this article talks about the necessity to be able to learn how to learn. Not just at the "adjust learning rate" level, but mechanisms that change the resulting coding, thereby creating different such contexts, or at least that's where I see the connection between those two. In essence: To get to AGI we need AIs which can develop their own topology.

    As to "rudeness": Make sure to never visit the Netherlands. Usually how this goes is that I link the article and the AI faithful I pointed it out to goes on a denial spree... because if they a) are actually into the topic, not just bystanders and b) did not have some psychological need to believe (including "my retirement savings are in AI stock") they c) would've come across the general argument themselves during their technological research. Or came up with it themselves, I've also seen examples of that: If you have a good intuition about complexity (and many programmers do) it's not unlikely a shower thought to have. Not as fleshed out as in the article, of course.

  • Leaked Documents Show OpenAI Has a Very Clear Definition of ‘AGI’

    Jump

  • I never knew that USB-C extensions are not allowed for a reason

    Jump

  • Why in the everloving would your electrical code allow sales of extension cords that can't withstand the whole of the plug/socket rating. If it's an adapter from a higher amperage plug to lower amperage socket you need a fuse.