SideQuark

6 ms·

91.
by SideQuark 6mo ago
The world has had many types of logic before and after Boolean logic was created, many used in computing. Boolean logic isn’t a constraint; it’s used where it’s useful, and others are used where they’re useful.
92.
by SideQuark 6mo ago
No, it approximates exp poorly over an infinitesimally small interval compared to exp. Resistors and capacitors are no where ideal components, which is why they have spec sheets to show how quickly they diverge. If we’re making sloppy appro
93.
by SideQuark 6mo ago
It’s only finite by putting the infinite series into an operation. And the basic monomial basis is not a single binary operation capable of reproducing the set of basic arithmetic ops. If you want trivial and basic, pick Peano postulates. B
94.
by SideQuark 6mo ago
Any transcendental function can be produced by arithmetic, since its complete for R. Go ahead and show how to compute exp or ln without an infinite series without circular reasoning. You can’t, since they’re transcendental. There are infini
95.
by SideQuark 6mo ago
Show me a way to physically compute exp or ln that is less gates than add. More gates means costlier in $, more energy in compute, and for these functions, higher latency. You don’t get to make up free ops, claim there is no cost in reality
96.
by SideQuark 6mo ago
Ooh, that 2nd link has a nice construction by Terry Tao giving a clear way to show infinitely many such functions exist for pretty much any set of operations.
97.
by SideQuark 6mo ago
Do you claim all things you don’t understand are LLMs? This is what I mean by these and many of your other comments being extremely poor quality to the point of deliberate ignorance. The paper above was published in 2012 [1], so that’s quit
98.
by SideQuark 6mo ago
Yes it can, by using the same infinite series that exp and ln use to compute. This one just costs less in money, hardware, energy, and is faster for basically every basic op.
99.
by SideQuark 6mo ago
The exp and ln are infinite series. Exp is roughly the infinite series for cos AND the infinite series for sin. Hiding that every op is an infinite series behind a name doesn’t make things free. It just makes even trivial ops like 1+2 vastl
100.
by SideQuark 6mo ago
And those come from the infinite series needed to compute exp and ln. They’re just as much work either way. The exp and ln way are vastly costlier for every op, including simply adding 1 and 2.
101.
by SideQuark 6mo ago
Computing exp or ln is an infinite series, and vastly more compute. Hiding series behind a name doesn’t make them free to compute.
102.
by SideQuark 6mo ago
It's math. You can check it yourself instead of this (and many other) thoughtless posts.
103.
by SideQuark 6mo ago
This isn't unique, or even the least compute way to do this. For example, let f(x,y) = 1/(x-y). This too is universal. I think there's a theorem stating for any finite set of binary operators there is a single one replacing i
104.
by SideQuark 6mo ago
The compute, energy, and physical cost of this versus a simple x+y is easily an order of magnitude. It will not replace anything in computing, except maybe fringe experiments.
105.
by SideQuark 6mo ago
This is not a 17000 qubit general computer. Read the paper.
106.
by SideQuark 6mo ago
They did not make a 17000 qubit computer. The qubits were not controllable or general in any way. The paper is linked, look at it. This title is misleading.
107.
by SideQuark 6mo ago
It ignores we already had that, in 2016, with experts from all over the world doing inspections and agreeing it worked. Then Trump blew up the deal against the wishes of the rest of the free world, claiming he’d make a better deal, which he
108.
by SideQuark 6mo ago
Those “few drones” have completely kept the US military, ships and all, far away since they can damage and sink large expensive vessels with tiny cheap drones. How did the planes and ships and missles fare in Iraq or Afghanistan? Oh yeah, d
109.
by SideQuark 6mo ago
The old govt was about to be toppled by people sick of it. The US attack unified those people behind the leaders son, someone they’d not have taken before, and entrenched a new generation against the US. So far the carrot and stick has them
110.
by SideQuark 6mo ago
How’d that plan work out in Iraq or Afghanistan, both much smaller, less armed countries? Decades and trillions spent, and what exactly did the US “win”?
111.
by SideQuark 6mo ago
Iran was little threat to the US before the US attacked. Now the US likely has earned itself more decades of terrorists, while simultaneously losing its military and political support from other countries. If the US objective was self destr
112.
by SideQuark 6mo ago
About 3000 citations, so I’d guess the paper was very well received
113.
by SideQuark 6mo ago
AI was used fundamentally for COVID vaccine development. AI is used for research in all modern drugs. It’s a certainty if cancer gets cured AI will have played a fundamental role since it’s already fundamental to precursors.
114.
by SideQuark 7mo ago
This is oft repeated nonsense you can disprove yourself. Make a reasonable basket of goods, write down current prices. Look up ads for those goods over time, write down prices. Compute inflation. Check BLS inflation. If there were even a ti
115.
by SideQuark 7mo ago
If they were taught what was representable and why they’d learn it quickly. And those that forget details later know to chase it down again if they need it. Making it voodoo hides that it’s learnable, deterministic, and useful to understand
116.
by SideQuark 7mo ago
Yes. A computer scientist should know how numbers are represented and not expect non-representable numbers in that format to be representable. 0.1 is just as non-representable in floating point as is pi as is 100^100 in a 32 bit integer. Te
117.
by SideQuark 7mo ago
There’s plenty of cases where ‘==‘ is correct. If you understand how floating point numbers work at the same depth you understand integers, then you may know the result of each side and know there’s zero error. Anything to do “approximately
118.
by SideQuark 7mo ago
Of all the valuable metrics on that site, all of which grok does badly at except one, you managed to pick that single one. https://artificialanalysis.ai/models
119.
by SideQuark 7mo ago
Removing 30% of the supply of a very important, and completely irreplaceable for most of its uses, resource isn’t hysteria.
120.
by SideQuark 7mo ago
The surplus had dried up before Bush’s election, as the dot com bust tanked federal tax revenues. Bush tax cuts did add to the deficit though.

More ›