Google: We've achieved quantum supremacy! IBM: Nope. And stop using that word, please The esoteric world of quantum computing is all aquiver following a robust blog post from IBM essentially rubbishing claims from Google that it has achieved "quantum supremacy". The post notes that quantum computing is approaching the limits of classical simulation and there are big questions as to how to evaluate and benchmark …
Apple. I bought one in the 1980s. Relevance?
No, seriously, whatever the pros and cons of IBM if it came down to a snap decision on who was right - the people who develop and sell advanced computers and software or the people who use them, no matter how cleverly, to support their advertising agency - I think I know where my vote would go.
Given that IBM has been pouring money into QC and that they don't make anywhere near as many classic computers as they did 30 or so years ago, I think you are on the right track but following the wrong reason.
It's all about undermining the competition. You wouldn't want that significant US government funding to go to someone without IBMs long history of pioneering QC...
IBM Research has generally been a solid organization. True, sometimes the marketing machine will hear of something and go on a tear. On the whole, though, they tend to publish solid work in the usual dry style.
This is the organization that invented the Scanning Tunneling Microscope, after all. Among other things.
There are two branches in Philosophy. One branch ended on Hegel, and I am its only living representative. The second is Marx, Moore, Bertrand Russell and Wittgenstein.
Modern computer, SQL, Google with Facebook came from Russell and Co., this is the philosophy of the constants, 1 and 0.
Quantum computer, AI are descended from Hegel and me, they are based on becoming, on what is between 1 and 0.
Otherwise, Russell and Co. are Arithmetic, Geometry and Algebra.
Quantum computer and AI are Differential Analysis and Lobachevsky Geometry.
Common sense?
I'm not saying IBM are wrong, but this is as much about publicity as being correct - if IBM had a 50+ qubit quantum computer that they could demonstrate, they would make the same claims rather than just announce one soon...
Interestingly, in critiquing their competition, I can't help wondering if they have actually revealed the limits of their own solutions and QC in general.
Is it a case of the closer we get to Quantum Supremacy the less we know about what it means? .... rsole
No, it is not, rsole. It is a case of the closer we get to Quantum Supremacy the more an elite and/or enlightening few know about what it means for everything.
And yes, that can be terrifyingly spun to terrorise any mass of natives captured in a collapsing corrupted system but whether that is to be effectively protective or quickly self-destructive is a mystery and enigma to exercise.
Touché, AC.
And that is a catastrophic vulnerability to exhaust, exploit and explore ...... adore and server, and thus also, coincidentally by SecureD IntelAIgent Design, an almighty opportunity to embrace, enjoy and employ in Advanced Ware Fighter Experiments.
:-) * ..For more 00mph for Investors in Sure Real Master Pilots into New JOINT AIdVentures. :-) I Kid U Not.:-)>
Methinks a Google claiming to have achieved quantum supremacy is a tad premature and untrue and thus utterly bogus .... for is not quantum communication on Earth just in its infancy .... and learning how to grow almightily well and wisely.
Which you might like to agree is a hell of a lot better than any other root route to be made available. :-) Capiche?
"In RL, a software agent takes sequential actions aiming tocmaximize a reward function, or a negative cost function, that embodies the target problem. Successful training of an RL agent depends on balancing exploration of unknown territory with exploitation of existing knowledge." https://www.nature.com/articles/s41534-019-0141-3.pdf
How does Google do THIS? Google works with patterns, each of which consists of several (two-five) words. Next, Google creates their absolute uniqueness by annotating each of their words with a set of some dictionary definitions, which convey their unique meanings. This uniqueness allows Google to instantly find the right patterns, without going through millions the same patterns in search for the right context: the context is literally "captured" as these unique dictionary definitions. So, "Yes", Google can do the work of many thousands of years almost instantaneously.
... it's a good blog post. The abstruse technical content is relatively minor and can safely be skipped by people not knowledgeable about the field. It makes good points about how data representations, algorithms, implementation optimizations, and resources with a range of economic parameters can be used to solve problems using conventional (non-BQP) techniques in ways that are many orders of magnitude more efficient than the straightforward approach.
Its cautions about "quantum supremacy" are also well-taken, of course. Even if (probably when) we have a true QS machine, the set of useful problems that can be economically solved by such a system is going to be quite small, even in the best case. Real, productive application of BQP algorithms will mostly require a machine quite a lot larger than the minimal QS system. And QC systems do not trivially scale. It's possible that a minimal-QS machine is feasible but a generally-useful QC system is not, even for high-value problems.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
