Willow, Our Quantum Chip

https://blog.google/technology/research/google-willow-quantum-chip/

by robflaherty • 2 years ago

1410 521 2 years ago

tombert• 2 years ago

Interesting; it might be time for me to load up a quantum simulator and star learning how to program these things.

I've pushed that off for a long time since I wasn't completely convinced that quantum computers actually worked, but I think I was wrong.

nick__m• 2 years ago

IBM has well structured learning material and a quantum simulator at https://learning.quantum.ibm.com/

Also my almamater made the quantum enigmas series that is appropriate for high-school students (it also interesting if you have no prior knowledge about quantum computing) https://www.usherbrooke.ca/iq/quantumenigmas/ (it also use IBM online learning platform)

gordon_freeman• 2 years ago

So one of the interesting comparisons between Quantum computing vs classical in the video: 5 mins vs 10^25 years. So are there any tradeoffs or specific cases in which the use cases for Quantum computing works or is this generic for "all" computing use cases? if later then this will change everything and would change the world.

dabeeeenster• 2 years ago

Its for a very, very, very narrow set of algorithms AFAIUI.

antognini• 2 years ago

There are only certain kinds of computing tasks which are amenable to an exponential speedup from quantum computing. For many classical algorithms the best you get from a quantum computer is an improvement by a factor of sqrt(N) by using Grover's algorithm.

The other tradeoff is that quantum computers are much noisier than classical computers. The error rate of classical computers is exceedingly low, to the extent that most programmers can go their entire career without even considering it as a possibility. But you can see from the figures in this post that even in a state of the art chip, the error rates are of order ~0.03--0.3%. Hopefully this will go down over time, but it's going to be a non-negligible aspect of quantum computing for the foreseeable future.

pitpatagain• 2 years ago

It is specific to cases where a quantum algorithm exists that provides speedup, it is not at all generic. The complexity class of interest is BQP: https://en.wikipedia.org/wiki/BQP

Also of note: P is in BQP, but it is not proven that BQP != P. Some problems like factoring have a known polynomial time algorithm, and the best known classical algorithm is exponential, which is where you see these massive speedups. But we don't know that there isn't an unknown polynomial time classical factoring algorithm and we just haven't discovered it yet. It is a (widely believed) conjecture, that there are hard problems solved in BQP that are outside P.

nuz• 2 years ago

Every time this comes up people say they're not actually useful for ML. Is that true? And if not what would they be useful for

christkv• 2 years ago

In principal NP complete problems is my guess.

wazdra• 2 years ago

It is unknown whether quantum computing makes NP-complete problems easier to solve. There is a complexity class for problems that can be solved "efficiently" on using quantum computing, called BQP. How BQP and NP are related is unknown. In particular, if an NP-complete problem was shown to be solvable efficiently with Quantum Computing (and thus in BQP), this open (and hard) research question would be solved (or at least half of it).

Note that BQP is not "efficient" in a real-word fashion, but for theoretical study of Quantum computing, it's a good first guess

benbayard• 2 years ago

AFAIK, which is not much, I believe it is problems that you can turn in to a cycle. Right now we pull out answers from quantum computers at random, but typically do not know what the inputs were that got that answer. But if you can get the answers from the quantum computer to be cyclical you can use that symmetry to get all the information you need.

whimsicalism8• 2 years ago

breaking crypto, for one

LampCharger• 2 years ago

No, a true quantum computer will not necessarily solve NP-complete (NPC) problems efficiently. Quantum algorithms like Grover’s provide quadratic speedups, but this is insufficient to turn exponential-time solutions into polynomial-time ones. While quantum computers excel in specific tasks (e.g., Shor’s algorithm for factoring), there’s no evidence they can solve all NP-complete problems efficiently.

Current complexity theory suggests that , the class of problems solvable by quantum computers, does not encompass . Quantum computers may aid in approximations or heuristics for NPC problems but won’t fundamentally resolve them in polynomial time unless , which remains unlikely.

vessenes7• 2 years ago

Factoring. Reversing ECC operations. Decrypting all the data thought to be safely stored at rest in any non quantum resistant storage.

I do think ai algorithms could be built that quantum gates could be fast at, but I don’t have any ideas off the top of my head this morning. If you think of AI training as searching the space of computational complexity and quantum algorithms as accessing a superposition of search states I would guess there’s an intersection. Google thinks so too - the lab is called quantum ai.

jawns• 2 years ago

It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse

I see the evidence, and I see the conclusion, but there's a lot of ellipses between the evidence and the conclusion.

Do quantum computing folks really think that we are borrowing capacity from other universes for these calculations?

gaze2• 2 years ago

I'm upset they put this in because this is absolutely not the view of most quantum foundations researchers.

whimsicalism8• 2 years ago

the everettian view is absolutely not the view? i am not so sure.

or you mean specifically the parallel computation view?

klipt• 2 years ago

Soon: "are alien universes slowing down your internet? Click here to learn more!"

Reminds me of the Aorist Rods from Hitchhikers' Guide to the Galaxy.

ColinHayhurst• 2 years ago

Credibility of the article plummeted when I got to that sentence, and especially since using name dropping.

ferfumarma• 2 years ago

One of the biggest problems with such an assertion is that it's not falsifiable.

It could be that we are borrowing qbit processing power from Russel's quantum teapot.

hshshshshsh8• 2 years ago

Science is not based on consensus seeking.

Science is about coming up with the best explanations irrespective of whether or not a large chunk does not believe it.

And best explanations are the ones that is hard to vary. Not the one that is most widely accepted or easy to accept based on the current world view.

justinpombrio• 2 years ago

From Wikipedia[1]:

A poll of 72 "leading quantum cosmologists and other quantum field theorists" conducted before 1991 by L. David Raub showed 58% agreement with "Yes, I think MWI is true".[85]

Max Tegmark reports the result of a "highly unscientific" poll taken at a 1997 quantum mechanics workshop. According to Tegmark, "The many worlds interpretation (MWI) scored second, comfortably ahead of the consistent histories and Bohm interpretations."[86]

In response to Sean M. Carroll's statement "As crazy as it sounds, most working physicists buy into the many-worlds theory",[87] Michael Nielsen counters: "at a quantum computing conference at Cambridge in 1998, a many-worlder surveyed the audience of approximately 200 people... Many-worlds did just fine, garnering support on a level comparable to, but somewhat below, Copenhagen and decoherence." But Nielsen notes that it seemed most attendees found it to be a waste of time: Peres "got a huge and sustained round of applause…when he got up at the end of the polling and asked 'And who here believes the laws of physics are decided by a democratic vote?'"[88]

A 2005 poll of fewer than 40 students and researchers taken after a course on the Interpretation of Quantum Mechanics at the Institute for Quantum Computing University of Waterloo found "Many Worlds (and decoherence)" to be the least favored.[89]

A 2011 poll of 33 participants at an Austrian conference on quantum foundations found 6 endorsed MWI, 8 "Information-based/information-theoretical", and 14 Copenhagen;[90] the authors remark that MWI received a similar percentage of votes as in Tegmark's 1997 poll.[90]

[1] https://en.wikipedia.org/wiki/Many-worlds_interpretation#Pol...

korkybuchek2• 2 years ago

Do quantum computing folks really think that we are borrowing capacity from other universes for these calculations?

Tangentially related, but there's a great Asimov book about this called The Gods Themselves (fiction).

vessenes7• 2 years ago

I’m partial to Anathem by Stephenson on this topic as well

wasabi99101124• 2 years ago

I was also really taken aback by this quote.

I have no idea who put it there, but I can assure you the actual paper contains no such nonsense.

I would have thought whoever writes the google tech blogs is more competent than bottom tier science journalists. But in this case I think it is more reasonable to assume malice, as the post is authored by the Google Quantum AI Lead, and makes more sense as hype-boosting buzzword bullshit than as an honest misunderstanding that was not caught during editing.

hshshshshsh8• 2 years ago

Quantum computation done in done multiple universes is the explanation given by David Deutsch the father of Quantum Computing. He invented the idea of a quantum computer to test the idea of parallel universes.

If you are okay with a single universe coming to existence out of nothing you should be able to handle parallel universes as well just fine.

Also your comment does not have any useful information. You assumed hype as the reason why they mentioned parallel computing. It's just a bias you have on looking at world. Hype does helps explain a lot of things. So it can be tempting to use it as a placeholder for anything that you don't accept based on your current set of beliefs.

movpasd2• 2 years ago

There are compelling arguments to believe in the many-worlds interpretation.

No sign of a Heisenberg cut has been observed so far, even as experiments involving entanglement of larger and larger molecules are performed, which makes objective-collapse theories hard to consider seriously.

Bohmian theories are nice, but require awkward adjustments to reconcile them with relativity. But more importantly, they are philosophically uneconomical, requiring many unobservable — even theoretically — entities [0].

That leaves either many-worlds or a quantum logic/quantum Bayesian interpretations as serious contenders [1]. These interpretations aren't crank fringe nonsense. They are almost inevitable outcomes of seriously considering the implications of the theory.

I will say that personally, I find many-worlds to focus excessively on the Schrödinger-picture pure state formulation of quantum mechanics. (At least to the level that I understood it — I expect there is literature on the connection with algebraic formulations, but I haven't taken the time to understand it.) So I would lean towards quantum logic–type interpretations myself.

The point of this comment was to say that many-worlds (or "multiverses", though I dislike the term) isn't nonsense. But it also isn't exactly the kind of sci-fi thing non-physicists might picture. Given how easy it is to misinterpret the term, however, I must agree with you that a self-aware science communicator would think twice about whether the term should be included, and that there may be not-so-scrupulous intentions at play here.

Quick edit: I realise the comment I've written is very technical. I'm happy to try to answer any questions. I should preface it by stating that I'm not a professional in the field, but I studied quantum information theory at a Masters level, and always found the philosophical questions of interest.

---

[0] Many people seem to believe that many-worlds also postulates the existence of unobservable parallel universes, but this isn't true. We observe the interaction of these universe's every time we observe quantum interference.

While we're here, we can clear up the misconception about "branching" — there is no branching in many-worlds, just the coherent evolution of the universal wave function. The many worlds are projections out of that wave function. They don't discretely separate from one another, either — it depends on your choice of basis. That choice is where decoherence comes in.

[1] And of course, there is the Copenhagen "interpretation" — preferred among physicists who would rather not think about philosophy. (A respectable choice.)

vixen99• 2 years ago

Presumably the 'nonsense' is the supposed link between the chip and MW theory.

Let me add a recommendation for David Wallace's book The Emergent Multiverse - a highly persuasive account of 'quantum theory according to the Everett Interpretation'. Aside from the technical chapters, much of it is comprehensible to non-physicists. It seems that adherents to MW do 'not know how to refute an incredulous stare'. (From a quotation)

ComputerGuru• 2 years ago

It doesn’t make sense to me because if we can borrow capacity to perform calculations then we can “borrow” an infinite amount of energy.

griomnib2• 2 years ago

Climate change solved: steal energy from adjacent universes, pipe our carbon waste into theirs.

rdtsc3• 2 years ago

It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch.

That's in line with a religious belief. One camp believes one thing, other believes something else, others refuse to participate and say "shut up and calculate". Nothing wrong with religious beliefs of course, it's just important to know that is what it is.

paxys2• 2 years ago

I don't understand the jump from: classical algorithm takes time A -> quantum algorithm takes time B -> (A - B) must be borrowed from a parallel universe.

Maybe A wasn't the most efficient algorithm for this universe to begin with?

kelnos6• 2 years ago

Right, and that's part of the argument against quantum computing being a proof (or disproof) of the many-worlds interpretation. Sure, "(A-B) was borrowed from parallel universes" is a possible explanation for why quantum computing can be so fast, but it's by far not the only possible explanation.

aithrowawaycomm3• 2 years ago

I suspect the real issue is that Big Tech investors and executives (including Sundar Pichai) are utterly hopped up on sci-fi, and this sort of stuff convinces them to dedicate resources to quantum computing.

kridsdale16• 2 years ago

That explains metaverse funding at least.

ko27• 2 years ago

It's a perfectly legit interpretation of what's happening, and many physicists share the same opinion. Of course the big caveat is that you need to interfere those worlds so that they cancel out, which necessarily requires a lower algorithmic bound which prevents you from doing infinite amount of computation in an instant.

johnfn• 2 years ago

You don't even have to get to the point where you're reading a post off Scott Aaronson's blog[1] at all; his headline says "If you take nothing else from this blog: quantum computers won't solve hard problems instantly by just trying all solutions in parallel."

[1] https://scottaaronson.blog/

aithrowawaycomm3• 2 years ago

In the same way people believe P != NP, most quantum computing people believe BQP != NP, and NP-complete problems will still take exponential time on quantum computers. But if we had access to arbitrary parallel universes then presumably that shouldn't be an issue.

The success on the random (quantum) circuit problem is really a valdiation of Feynman's idea, not Deutsch: classical computers need 2^n bits to simulate n qubits, so we will need quantum computers to efficiently simulate quantum phenomena.

jumping_frog• 2 years ago

Does access to arbitrary parallel universes imply that they divide up the computation and the correct answer is distributed to all of the universes or in such a collection, there will be sucker universes which will always receive wrong answers ?

melvinmelih• 2 years ago

It performed a computation in under five minutes that would take one of today’s fastest supercomputers 1025 or 10 septillion years. If you want to write it out, it’s 10,000,000,000,000,000,000,000,000 years.

If it's not, what would be your explanation for this significant improvement then?

Ar-Curunir2• 2 years ago

I mean, that's like saying GPUs operate in parallel universes because they can do certain things thousands of times faster than CPUs.

shawabawa32• 2 years ago

Quantum computing can perform certain calculations much faster than classical computing in the same way classical computing can perform certain calculations much faster than an abacus

GenerWork• 2 years ago

Do quantum computing folks really think that we are borrowing capacity from other universes for these calculations?

Doesn't this also mean that other universes have civilizations that could potentially borrow capacity from our universe, and if so, what would that look like?

hshshshshsh8• 2 years ago

The quantum computer idea was literally invented by David Deutsche to test the many universes theory of quantum physics.

wasabi99101124• 2 years ago

You've mentioned this in another comment. I have to point out, even if this is his opinion, and he has been influential in the field, it does not mean that this specific idea of his has been influential.

Vegenoid• 2 years ago

It is not useful to spam this comment repeatedly under different people who question or disagree with many worlds. Pick one place to make your case.

killerstorm• 2 years ago

Everett interpretation simply asserts that quantum wavefunctions are real and there's no such thing as "wavefunction collapse". It's the simplest interpretation.

People call it "many worlds" because we can interact only with a tiny fraction of the wavefunction at a time, i.e. other "branches" which are practically out of reach might be considered "parallel universes".

But it would be more correct to say that it's just one universe which is much more complex than what it looks like to our eyes. Quantum computers are able to tap into this complexity. They make a more complete use of the universe we are in.

tobias2014• 2 years ago

This might turn into a debate of defining "simplest", but I think the ensemble/statistical interpretation is really the most minimal in terms of fancy ideas or concepts like "wavefunction collapse" or "multiverses". It doesn't need a wavefunction collapse nor does it need multiverses.

layer83• 2 years ago

The Schrödinger equation inherently contains a multiverse. The disagreement is about whether the wave function described by the equation collapses to a single universe upon measurement (i.e. whether the equation stops holding upon measurement), or whether the different branches continue to exist (i.e. the equation continues to hold at all times), each with a different measurement outcome. Regardless, between measurements the different branches exist in parallel. It’s what allows quantum computation to be a thing.

Lionga• 2 years ago

Non of that honkey ponkey is needed if you just give up locality and a hard deterministic explanation like De-Broglie-Bohm gives all the same correct measurements and conclusions like Copenhagen interpretation without multiverses and "wave function collapses".

Copenhagen interpretation is just "easier" (like oops all our calculations about the univers don't seemt to fit, lets invent "dark matter") when the correct explanations makes any real world calculation practically impossible (thus ending most of physics further study) as any atom depends on every other atom at any time.

wasabi99101124• 2 years ago

The Schrödinger equation inherently contains a multiverse.

A simple counterexample is superdeterminism, in which the different measurement outcomes are an illusion and instead there is always a single pre-determined measurement outcome. Note that this does not violate Bell's inequality for hidden variable theories of quantum mechanics, as Bell's inequality only applies to hidden variables uncorrelated to the choice of measurement: in superdeterminism, both are predetermined so perfectly correlated.

creata• 2 years ago

The Schrödinger equation inherently contains a multiverse.

Just to be clear, where in the Schrödinger equation (iħψ̇ = Hψ) is the "multiverse"?

varjag2• 2 years ago

Yes this is deeply unserious tangent in supposedly landmark technology announcement.

athesyn• 2 years ago

It's just marketing.

qnleigh10• 2 years ago

This is a viable interpretation of quantum mechanics, but currently there is no way to scientifically falsify or confirm any particular interpretation. The boundary between philosophy and science is fuzzy at times, but this question is solidly on the side of philosophy.

That being said, I think the two most commonly preferred interpretations of quantum mechanics among physicists are 'Many Worlds' and 'I try not to think about it too hard.'

m3kw9• 2 years ago

So are we now concerned with the environment of another universe? Like climate activitist but for multiverses?

whoitwas2• 2 years ago

Well. If you study quantum physics and the folks who found it like Max Planck, they believed in "a conscious and intelligent non-visible living energy force .. . the matrix mind of all matter".

I don't know much about multiverse, but we need something external to explain the magic we uncover.

Energy and quantum mechanics are really cool but dense to get into. Like Planck, I suspect there's a link between consciousness and matter. I also think our energy doesn't cease to exist when our human carcass expires.

htrp• 2 years ago

105 qubits

wslh• 2 years ago

ELI5: what I could do if I have this chip at home?

wrsh07• 2 years ago

Probably just research on quantum computers? I don't think it's big enough to let you solve any practical problems, but maybe someone can correct me

d3m0t3p• 2 years ago

IF (and that's a big if) that's true then it means they can factorize number into primes with this quantum computer and break encryption.

readyplayernull• 2 years ago

It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch.

Processing in multiverse. Would that mean we are inyecting entropy into those other verses? Could we calculate how many are there from the time it takes to do a given calculation? We need to cool the quantum chip in our universe, how are the (n-1)verses cooling on their end?

deanCommie3• 2 years ago

What if we are? And by injecting entropy into it, we are actually hurrying (in small insignificant ways) the heat death of those universes? What if we keep going and scale out and in the future it causes a meaningful impact to that universe in a way that it's residents would be extremely unhappy with, and would want to take revenge?

What if it's already happening to our universe? And that is what black holes are? Or other cosmology concepts we don't understand?

Maybe a great filter is your inability to protect your universe from quantum technology from elsewhere in the multiverse ripping yours up?

Maybe the future of sentience isn't fighting for resources on a finite planet, or consuming the energy of stars, but fighting against other multiverses.

Maybe The Dark Forest Defence is a decision to isolate your universe from the multiverse - destroying it's ability to participate in quantum computation, but also extending it's lifespan.

(I don't believe ANY of this, but I'm just noting the fascinating science fiction storylines available)

navaati• 2 years ago

Getting strong vibes of Asimov’s novel "The Gods Themselves" here ! For those who haven’t read it I recommend it. It’s a nice little self-contained book, not a grandiose series and universe, but I love it.

kridsdale16• 2 years ago

I’d say it’s more akin to Dark Energy than anything Black Hole related.

DE is some sort of entropy that is being added to our cosmos in an exponential way over historic time. It began at a point a few billion in to our history.

Willish42• 2 years ago

Thanks for throwing in references like https://en.wikipedia.org/wiki/Dark_forest_hypothesis even though this was a silly response to the science fiction implications.

I found it an interesting read and hadn't heard the term before, but it's exactly the kind of nerdy serendipity I come to this site for!

jsvlrtmred• 2 years ago

AFAIK a fundamental step in any quantum computing algorithm is bringing the qubits back to a state with a nonrandom outcome (specifically, the answer to the problem being solved). Thus a "good" quantum computer does not bifurcate the wavefunction at a macro level, ie there is no splitting of the "multiverse" after the calculation.

thrance3• 2 years ago

The many-worlds interpretation of quantum theory [1] is widely considered unfalsifiable and therefore mostly pseudoscientific. This article is way in over it's head in claiming such nonsense.

[1] https://en.wikipedia.org/wiki/Many-worlds_interpretation

zarzavat• 2 years ago

That depends on whether you think that MWI is making claims beyond what QM claims, or if you think that MWI is just a restatement of QM from another perspective. If the latter then the falsifiability of MWI is exactly the same as the falsifiability of QM itself.

newsbinator• 2 years ago

Unfalsifiable does not mean pseudoscientific. Plenty of things might be unfalsifiable for humans (e.g. humans lack the cognitive capacity to conceive of ways to falsify them), but then easily falsifiable for whatever comes after humans.

akira2501• 2 years ago

Isn't it really just processing with matrices of imaginary numbers? Which is why you need error correction in the first place because the higher temperature the system the less coherent the phases all become? Thus having absolutely nothing to do with multiverse theory?

I think string theories ideas about extra curled up dimensions are far more likely places to look. You've already got an infinite instantaneous energy problem with multiverses let alone entropy transfer considerations.

ipsum2• 2 years ago

They renamed quantum supremacy to "beyond-classical"? That's something.

maxboone• 2 years ago

https://arxiv.org/abs/1705.06768

It's not something that new, I like it.

sdenton4• 2 years ago

Quantum supremacy was an absolutely awful name for what it was (ability to do something, anything, better than a classical computer, which remains 'supreme' on all problems of any practical interest).

airstrike• 2 years ago

Sure, but it sounded much cooler

Mistletoe• 2 years ago

In what ways could Google monetize quantum computing?

michaelt• 2 years ago

Searching through an unstructured data set of size N on a classical computer takes O(N) time

but on a quantum computer, Grover's Algorithm allows such a search to be performed in O(N^0.5) time.

So Quantum Computing, could bring us a future where, when you perform a Google search for a word, the web pages returned actually contain the word you searched for.

_benj2• 2 years ago

So Quantum Computing, could bring us a future where, when you perform a Google search for a word, the web pages returned actually contain the word you searched for.

Lol! I'm not gonna put a kagi plug here...

7e• 2 years ago

Quantum mechanics is a computational shortcut that makes our simulation cost-effective. Mass adoption of chips like these is going to make the particular situation we live in unprofitable for hosts, resulting in the firey and dramatic end of the world for us. Simulating ancestors is fun, but not after your cloud bill skyrockets. Thank you, Google, for bringing about the apocalypse.

vessenes7• 2 years ago

I’m a quantum dabbler so I’ll throw out an armchair reaction: this is a significant announcement.

My memory is that 256 bit keys in non quantum resistant algos need something like 2500 qubits or so; and by that I mean generally useful programmable qubits. To show a bit over 100 qubits with stability, meaning the information survives a while, long enough to be read, and general enough to run some benchmarks on is something many people thought might never come.

There’s a sort of religious reaction people have to quantum computing: it breaks so many things that I think a lot of people just like to assume it won’t happen: too much in computing and data security will change -> let’s not worry about it.

Combined with the slow pace of physical research progress (Schorrs algorithm for quantum factoring was mid 90s), and snake oil sales companies, it’s easy to ignore.

Anyway seems like the clock might be ticking; AI and data security will be unalterably different if so. Worth spending a little time doing some long tail strategizing I’d say.

goatking• 2 years ago

How can I, a regular software engineer, learn about quantum computing without having to learn quantum theory?

Worth spending a little time doing some long tail strategizing I’d say

any tips for starters?

sshb• 2 years ago

Might be worth checking out: https://quantum.country/

zitterbewegung2• 2 years ago

I recommend this book I studied it in Undergrad and I never took a quantum theory course. https://www.amazon.com/Quantum-Computing-Computer-Scientists...

potsandpans• 2 years ago

Start here: https://youtu.be/F_Riqjdh2oM

You don't need to know quantum theory necessarily, but you will need to know some maths. Specifically linear algebra.

There are a few youtube courses on linear algebra

For a casual set of video: - https://youtube.com/playlist?list=PLZHQObOWTQDPD3MizzM2xVFit...

For a more formal approach:

- https://youtube.com/playlist?list=PL49CF3715CB9EF31D

And the corresponding open courseware

- https://ocw.mit.edu/courses/18-06-linear-algebra-spring-2010...

Linear algebra done right comes highly recommended

- https://linear.axler.net/

jvanderbot• 2 years ago

There is a course mentioned in the article, but I'm not clear on how "theory" it is.

https://coursera.org/learn/quantum-error-correction

carabiner• 2 years ago

First learn about eigenvalues.

billti• 2 years ago

If you're a software engineer, then the Quantum Katas might fit your learning style. The exercises use Q#, which is quantum specific programming language.

https://quantum.microsoft.com/en-us/tools/quantum-katas

The first few lessons do cover complex numbers and linear algebra, so skip ahead if you want to get straight to the 'quantum' coding, but there's really no escaping the math if you really want to learn quantum.

Disclaimer: I work in the Azure Quantum team on our Quantum Development Kit (https://github.com/microsoft/qsharp) - including Q#, the Katas, and our VS Code extension. Happy to answer any other questions on it.

kvathupo2• 2 years ago

The bar for entry is surprisingly low, you just need to brush up on intro abstract algebra. I recommend the following:

1. Kaye, LaFlamme, and Mosca - An Introduction to Quantum Computing

2. Nielsen and Chuang - Quantum Computation and Quantum Information (The Standard reference source)

3. Andrew Childs's notes here [1]. Closest to the state-of-the-art, at least circa ~3 years ago.

[1] https://www.cs.umd.edu/~amchilds/qa/

neoden• 2 years ago

These lessons might be of help: https://youtu.be/3-c4xJa7Flk?si=krrpXMKh3X5ktrzT

currymj• 2 years ago

specifically avoid resources written by and for physicists.

the model of quantum mechanics, if you can afford to ignore any real-world physical system and just deal with abstract |0>, |1> qubits, is relatively easy. (this is really funny given how incredibly difficult actual quantum physics can be.)

you have to learn basic linear algebra with complex numbers (can safely ignore anything really gnarly).

then you learn how to express Boolean circuits in terms of different matrix multiplications, to capture classical computation in this model. This should be pretty easy if you have a software engineer's grasp of Boolean logic.

Then you can learn basic ideas about entanglement, and a few of the weird quantum tricks that make algorithms like Shor and Grover search work. Shor's algorithm may be a little mathematically tough.

realistically you probably will never need to know how to program a quantum computer even if they become practical and successful. applications are powerful but very limited.

"What You Shouldn't Know About Quantum Computers" is a good non-mathematical read.

https://arxiv.org/abs/2405.15838

ajb• 2 years ago

The simplest algorithm to understand is probably Grover's algorithm. Knowing that shows you how to get an sqrt(N) speedup on many classical algorithms. Then have a look at shor's algorithm which is the classic factoring algorithm.

I would not worry about hardware at first. But if you are interested and like physics, the simplest to understand are linear optical quantum circuits. These use components which may be familiar from high school or undergraduate physics. The catch is that the space (and component count) is exponential in the number of qubits, hence the need for more exotic designs.

cevi• 2 years ago

I always recommend Watrous's lecture notes: https://cs.uwaterloo.ca/~watrous/QC-notes/QC-notes.pdf

I prefer his explanation to most other explanations because he starts, right away, with an analogy to ordinary probabilities. It's easy to understand how linear algebra is related to probability (a random combination of two outcomes is described by linearly combining them), so the fact that we represent random states by vectors is not surprising at all. His explanation of the Dirac bra-ket notation is also extremely well executed. My only quibble is that he doesn't introduce density matrices (which in my mind are the correct way to understand quantum states) until halfway through the notes.

tsimionescu8• 2 years ago

If you want to learn about what theoretical quantum computers might be able to do faster than classical ones and what they might not, you can try to read about quantum complexity theory, or some of what Scott Aaronson puts out on his blog if you don't want to go that in depth.

But the key thing to know about quantum computing is that it is all about the mathematical properties of quantum physics, such as the way complex probabilities work.

thrance3• 2 years ago

The error rates given are still horrendous and nowhere near low enough for the Quantum Fourier Transform used by Shor's algorithm. Taking qubit connectivity into account, a single CX between 2 qubits that are 10 edges aways gives an error rate of 1.5%.

Also, the more qubits you have/the more instructions are in your program, the faster the quantum state collapses. Exponentially so. Qubit connectivity is still ridiculously low (~3) and does not seem to be improving at all.

About AI, what algorithm(s) do you think might have an edge over classical supercomputers in the next 30 years? I'm really curious, because to me it's all (quantum) snake oil.

LeftHandPath5• 2 years ago

Re: AI, it's a long way off still. The big limitation to anything quantum is always going to be decoherence and t-time [0]. To do anything with ML, you'll need whole circuit (more complex than shor's) just to initialize the data on the quantum device; the algorithms to do this are complex (exponential) [1]. So, you have to run a very expensive data-initialization circuit, and only then can you start to run your ML circuit. All of this needs to be done within the machine's t-time limit. If you exceed that limit, then the measured state of a qubit will have more to do with outside-world interactions than interactions with your quantum gates.

Google's willow chip has t-times of about 60-100mu.s. That's not an impressive figure -- in 2022, IBM announced their Eagle chip with t-times of around 400mu.s [2]. Google's angle here would be the error correction (EC).

The following portion from Google's announcement seems most important:

With 105 qubits, Willow now has best-in-class performance across the two system benchmarks discussed above: quantum error correction and random circuit sampling. Such algorithmic benchmarks are the best way to measure overall chip performance. Other more specific performance metrics are also important; for example, our T1 times, which measure how long qubits can retain an excitation — the key quantum computational resource — are now approaching 100 µs (microseconds). This is an impressive ~5x improvement over our previous generation of chips.

Again, as they lead with, their focus here is on error correction. I'm not sure how their results compare to competitors, but it sounds like they consider that to be the biggest win of the project. The RCS metric is interesting, but RCS has no (known) practical applications (though it is a common benchmark). Their T-times are an improvement over older Google chips, but not industry-leading.

I'm curious if EC can mitigate the sub-par decoherence times.

[0] https://www.science.org/doi/abs/10.1126/science.270.5242.163...

[1] https://dl.acm.org/doi/abs/10.5555/3511065.3511068

[2] https://www.ibm.com/quantum/blog/eagle-quantum-processor-per...

AlexCoventry• 2 years ago

In addition to that, the absolutely enormous domains that the Fourier Transform sums over (essentially, one term in the sum for each possible answer), and the cancellations which would have to occur for that sum to be informative, means that a theoretically-capable Quantum Computer will be testing the predictions of Quantum Mechanics to a degree of precision hundreds of orders of magnitude greater than any physics experiment to date. (Or at least dozens of orders of magnitude, in the case of breaking Discrete Log on an Elliptic Curve.) It demands higher accuracy in the probability distributions predicted by QM than could be confirmed by naive frequency tests which used the entire lifetime of the entire universe as their laboratory!

Imagine a device conceived in the 17th century, the intended functionality of which would require a physical sphere which matches a perfect, ideal, geometric sphere in Euclidean space to thousands of digits of precision. We now know that the concept of such a perfect physical sphere is incoherent with modern physics in a variety of ways (e.g., atomic basis of matter, background gravitational waves.) I strongly suspect that the cancellations required for the Fourier Transform in Shor's algorithm to be cryptographically relevant will turn out to be the moral equivalent of that perfect sphere.

We'll probably learn some new physics in the process of trying to build a Quantum Computer, but I highly doubt that we'll learn each others' secrets.

sesm3• 2 years ago

Also, the more qubits you have/the more instructions are in your program, the faster the quantum state collapses.

Was this actually measured and published somewhere?

JanisErdmanis8• 2 years ago

The required number of qubits to execute Shor’s algorithm is way larger than 2500 qubits as the error ceiling for logical qubits must decrease exponentially with every logical qubit added to produce meaningful results. Hence, repeated applications of error correction or an increase in the surface code would be required. That would significantly blow up the number of physical qubits needed.

adastra227• 2 years ago

He’s quoting the number of logical qubits (which is 1024 IIRC, not 2500), after error correction.

ETA: Wikipedia 2330 qubits, but I'm not sure it is citing the most recent work: https://en.wikipedia.org/wiki/Elliptic-curve_cryptography#ci...

The5thElephant• 2 years ago

Isn't that what they are claiming is true now? That the errors do decrease exponentially with each qubit added?

JanisErdmanis8• 2 years ago

Well, this is embarrassing. I just realised I had wrongly interpreted the result in [1]. I made an error on how Shor's algorithm encodes the numbers, wrongly assuming that numbers are encoded into quantum state space, which is 2^(2^n), where instead, there is one bit encoded into one qubit, which is also more practical.

The result shall be interpreted directly with the error rate for logical qubits to decrease as ~n^(-1/3). This, in turn, means that factorisation of a 10000-bit number would only require an error rate of 1/10th of the number of the logical qubits for a 10-bit number. This is practical given that one can make a quantum computer with around 100k qubits and correct errors on them.

On the other hand, a sibling comment already mentioned the limited connectivity that those quantum computers now have. This, in turn, requires a repeated application of SWAP gates to get the interaction one needs. I guess this would add a linear overhead for the noise; hence, the scaling of the error rate for logic qubits is around ~n^(-4/3). This, in turn, makes 10000-bit factorisation require a logical error rate of 1/10000 that of 10-bit number factorisation. Assuming that 10 physical qubits are used to reduce error by order, it can result in around 400k physical qubits.

[1] https://link.springer.com/article/10.1007/s11432-023-3961-3

isoprophlex• 2 years ago

Data security okay. But AI? How will that change?

adastra227• 2 years ago

AI is essentially search. Quantum computers are really good at search.

numpad02• 2 years ago

Aren't quantum computers expected to be like digitally read analog computers for high dimension optimization problems, and AI is like massive high dimension optimization problems?

winwang2• 2 years ago

Edit after skimming arxiv preprint[1]:

Yeah, this is pretty huge. They achieved the result with surface codes, which are general ECCs. The repetition code was used to further probe quantum ECC floor. "Just POC" likely doesn't do it justice.

(Original comment):

Also quantum dabbler (coincidentally dabbled in bitflip quantum error correction research). Skimmed the post/research blog. I believe the key point is the scaling of error correction via repetition codes, would love someone else's viewpoint.

Slightly concerning quote[2]:

"""

By running experiments with repetition codes and ignoring other error types, we achieve lower encoded error rates while employing many of the same error correction principles as the surface code. The repetition code acts as an advance scout for checking whether error correction will work all the way down to the near-perfect encoded error rates we’ll ultimately need.

"""

I'm getting the feeling that this is more about proof-of-concept, rather than near-practicality, but this is certainly one fantastic POC if true.

[1] https://arxiv.org/abs/2408.13687

[2] https://research.google/blog/making-quantum-error-correction...

Relevant quote from preprint (end of section 1, sorry for copy-paste artifacts):

"""

In this work, we realize surface codes operating below threshold on two superconducting processors. Using a 72-qubit processor, we implement a distance-5 surface code operating with an integrated real-time decoder. In addition, using a 105-qubit processor with similar performance, we realize a distance-7 surface code. These processors demonstrate Λ > 2 up to distance-5 and distance7, respectively. Our distance-5 quantum memories are beyond break-even, with distance-7 preserving quantum information for more than twice as long as its best constituent physical qubit. To identify possible logical error f loors, we also implement high-distance repetition codes on the 72-qubit processor, with error rates that are dominated by correlated error events occurring once an hour. These errors, whose origins are not yet understood, set a current error floor of 10−10. Finally, we show that we can maintain below-threshold operation on the 72qubit processor even when decoding in real time, meeting the strict timing requirements imposed by the processor’s fast 1.1µs cycle duration.

"""

echelon• 2 years ago

Google could put themselves and everyone else out of business if the algorithms that underpin our ability to do e-commerce and financial transactions can be defeated.

Goodbye not just to Bitcoin, but also Visa, Stripe, Amazon shopping, ...

wasabi99101124• 2 years ago

You got the main idea, it's a proof-of-concept: that a class of error-correcting code on real physical quantum chips obey the threshold theorem, as is expected based on theory and simulations.

However the main scaling of error correction is via surface codes, not repetition codes. It's an important point as surface codes correct all Pauli errors, not just either bit-flips or phase-flips.

They use repetition codes as a diagnostic method in this paper more than anything, it is not the main result.

In particular, I interpret the quote you used as: "We want to scale surface codes even more, and if we were able to do the same scaling with surface codes as we are able to do with repetition codes, then this is the behaviour we would expect."

Edit: Welp, saw your edit, you came to the same conclusion yourself in the time it took me to write my comment.

bee_rider• 2 years ago

I think some element of it might be: Shor’s algorithm has been known of for 30 years, and hypothetically could be used to decrypt captured communications, right? So, retroactively I will have been dumb for not having switched to a quantum-resistant scheme. And, dumb in a way that a bunch of academic nerds have been pointing out for decades.

That level of embarrassment is frankly difficult to face. And it would be devastating to the self-image of a bunch of “practical” security gurus.

Therefore any progress must be an illusion. In the real world, the threats are predictable and mistakes don’t slowly snowball into a crisis. See also, infrastructure.

adastra227• 2 years ago

What would you switch to? There hasn’t been post quantum systems to use until very very recently.

bawolff4• 2 years ago

All you encrypted communication from the 90s (SSL anyways) can probably be decrypted with classical means anyways. 90s SSL was pretty bad.

cherryteastain• 2 years ago

AI and data security will be unalterably different if so

Definitely agree with the latter, but do you have any sources on how quantum comphters make "AI" (i.e. matrix multiplication) faster?

meta_x_ai• 2 years ago

Exploring via Search can become O(1) instead of M^N

npalli• 2 years ago

Worth spending a little time doing some long tail strategizing I’d say.

Yup, like Bitcoin going to zero.

drcode• 2 years ago

eh, they will add a quantum-resistant signature scheme (already a well-understood thing) then people can transfer their funds to the new addresses before it is viable to crack the existing addresses

K0balt3• 2 years ago

I think you’re going to need about 10,000,000 qbits to divert a transaction, but that’s still within foreseeable scale. I think it’s extreme likely that the foundation will have finished their quantum resistance planning before we get to 10MM coherent qbits, but still, it’s a potential scenario.

More likely that other critical infrastructure failures will happen within trad-finance, much larger vulnerability footprint, and being able to trivially reverse engineer every logged SSL session is likely to be a much more impactful turn of events. I’d venture that there are significant ear-on-the-wire efforts going on right now in anticipation of a reasonable bulk SSL de cloaking solution. Right now we think it doesn’t matter who can see our “secure” traffic. I think that is going to change, retroactively, in a big way.

sekai• 2 years ago

Yup, like Bitcoin going to zero.

If the encryption on Bitcoin is broken, say goodbye to the banking system.

vessenes7• 2 years ago

I'm a little more in my wheelhouse here -- without an algo change, Grover's algorithm would privilege quantum miners significantly, but not any more than the industry has seen in the last 13 years (C code on CPU -> GPU -> Large Geometry ASIC -> Small Geometry ASIC are similarly large shifts in economics for miners probably).

As to faking signatures and, e.g. stealing Satoshi's coins or just fucking up the network with fake transactions that verify, there is some concern and there are some attack vectors that work well if you have a large, fast quantum computer and want to ninja in. Essentially you need something that can crack a 256 bit ECDSA key before a block that includes a recently released public key can be inverted. That's definitely out of the reach of anyone right now, much less persistent threat actors, much less hacker hobbyists.

But it won't always be. The current state of the art plan would be to transition to a quantum-resistant UTXO format, and I would imagine, knowing how Bitcoin has managed itself so far, that will be a well-considered, very safe, multi-year process, and it will happen with plenty of time.

m101• 2 years ago

Bitcoin will just fork to a quantum proof encryption scheme and there will be something called "bitcoin classic" that is the old protocol (which few would care about)

qnleigh10• 2 years ago

You need to distinguish between "physical qubits" and "logical qubits." This paper creates a single "first-of-a-kind" logical qubit with about 100 physical qubits (using Surface Code quantum error correction). A paper from Google in 2019 estimates needing ~20 million physical qubits ("How to factor 2048 bit RSA integers in 8 hours using 20 million noisy qubits" - https://arxiv.org/abs/1905.09749), though recent advances probably brought this number down a bit. That's because to run Shor's algorithm at a useful scale, you need a few thousand very high quality logical qubits.

So despite this significant progress, it's probably a still a while until RSA is put out of the job. That being said, quantum computers would be able to retroactively break any public keys that were stored, so there's a case to be made for switching to quantum-resistant cryptography (like lattice-based cryptography) sooner rather than later.

rhubarbtree2• 2 years ago

This is correct. I worked in quantum research a little.

rhaps0dy• 2 years ago

Thank you for the explanation. It's still an upwards update on the qubit timelines of https://arxiv.org/pdf/2009.05045 (see Fig. 7), but not by an insane amount. We've realized their 95% expectation of qubit progress (1 logical qubit) for 2026, in 2024.92 instead.

Which to be clear is quite a bit faster than expected in 2020, but still within the realm of plausible stuff.

kortilla4• 2 years ago

quantum computers would be able to retroactively break any public keys that were stored

Use a key exchange that offers perfect forward secrecy (e.g. diffie Hellman) and you don’t need to worry about your RSA private key eventually being discovered.

prmoustache• 2 years ago

so there's a case to be made for switching to quantum-resistant cryptography (like lattice-based cryptography) sooner rather than later.

This.

People seems to think that because something is end to end encrypted it is secure. They don't seem to grasp that the traffic and communication that is possibly dumped/recorded now in encrypted form could be used against them decades later.

r33b33• 2 years ago

Worth spending a little time doing some long tail strategizing I’d say.

What do you mean by this?

unethical_ban• 2 years ago

The only thing I know or understand about quantum computing is its ability to "crack" traditional encryption algorithms.

So the commenter is saying that Cybersecurity needs to be planning for a near-world where traditional cryptography, including lots of existing data at rest, is suddenly as insecure as plaintext.

tsimionescu8• 2 years ago

"long tail" typically refers to the tail of a normal distribution - basically it's a sciencey, but common, way of saying "very unlikely event". So, the OP was saying that it's worth spending some time strategizing about the unlikely event that a practical RSA-breaking QC appears in the near future, even though it's still a "long tail" (very unlikely) event.

Honestly, there's not that much to discuss on this though. The only things you can do from this strategizing is to consider even encrypted data as not safe to store, unless you're using quantum resistant encryption such as AES; and to budget time for switching to PQC as it becomes available.

sizzle2• 2 years ago

How long until this can derive a private key from its public key in the cryptocurrency space? Is this an existential threat to crypto?

bawolff4• 2 years ago

Long enough you don't need to panic or worry.

Short enough that its reasonable to start r&d efforts on post quantum crypto.

ryankshaw• 2 years ago

I had the same question and this article was really helpful in explaining the threat models

https://www.deloitte.com/nl/en/services/risk-advisory/perspe...

valval• 2 years ago

This is the same as climate change. Something might happen sometime in the future.

beams_of_light• 2 years ago

https://en.wikipedia.org/wiki/BGP_hijacking#Public_incidents

A long-term tactic of our adversaries is to capture network traffic for later decryption. The secrets in the mass of packets China assumedly has in storage, waiting for quantum tech, is a treasure trove that could lead to crucial state, corporate, and financial secrets being used against us or made public.

AI being able to leverage quantum processing power is a threat we can't even fathom right now.

Our world is going to change.

inasio• 2 years ago

They showed a logical qubit that can last entangled for an hour, but to do that they had to combine their hundred or so physical qubits into a single one, so in some sense they have, right now, a single (logical) qubit

weatherlite• 2 years ago

AI and data security will be unalterably different if so

So what are the implications if so ?

vhiremath42• 2 years ago

I really wish the release videos made things a ~tad~ bit less technical. I know quantum computers are still very early so the target audience is technical for this kind of release, but I can’t help wonder how many more people would be excited and pulled in if they made the main release video more approachable.

tcgv• 2 years ago

If you have programming experience, you might find this interesting: back in 2019, when Google announced achieving quantum supremacy, I worked on a personal project to study the basics of quantum computing and share my learnings with others in my blog:

- https://thomasvilhena.com/2019/11/quantum-computing-for-prog...

vhiremath42• 2 years ago

Excellent thank you!

rdtsc3• 2 years ago

The main part for me is reducing error faster as they scale. This was a major road-block, known as "below threshold”. That's a major achievement.

I am not sure about RCS as the benchmark as not sure how useful that is in practice. It just produced really nice numbers. If I had a few billions of pocket change around, would I buy this to run RCS really fast? -Nah, probably not. I'll get more excited when they factor numbers at a rate that would break public key crypto. For that would spend my pocket change!

vessenes7• 2 years ago

The implication seems to be that they can implement other gates. As my gen z kids say: huge if true.

wasabi99101124• 2 years ago

It's really important to note that the error correction test and the random circuit test are separate tests.

The error correction is producing a single logical qubit of quantum memory, i.e. a single qubit with no gates applied to it.

Meanwhile, the random circuit sampling uses physical qubits with no error correction, and is used as a good benchmark in part because it can prove "quantumness" even in the presence of noise.[1]

[1] https://research.google/blog/validating-random-circuit-sampl...

fidotron2• 2 years ago

The slightly mind blowing bit is detailed here: > https://research.google/blog/making-quantum-error-correction...

“the first quantum processor where error-corrected qubits get exponentially better as they get bigger”

Achieving this turns the normal problem of scaling quantum computation upside down.

thrance3• 2 years ago

It also breaks a fundamental law of quantum theory, that the bigger a system in a quantum state is, the faster it collapses, exponentially so. Which should at least tell you to take Google's announcement with z grain of salt.

wasabi99101124• 2 years ago

This is not a "fundamental law of quantum theory", as evidenced by the field of quantum error correcting codes.

Google's announcement is legit, and is in line with what theory and simulations expect.

bgnn• 2 years ago

Hmm why? I thought the whole idea was this would work eventually. Physical qubit vs logical qubit distinction is there already for a long time.

The scaling problem is multifaceted. IMHO the physical qubits are the biggest barrier to scaling.

fidotron2• 2 years ago

Hmm why?

In theory, theory and practice are the same.

whimsicalism8• 2 years ago

We need to seriously think if our systems/society are even remotely ready for this.

germandiago• 2 years ago

In order to evolve, forbiding evolution is the wrong path. Just use and study and learn from new things and accumulate experience is the way to go.

kridsdale16• 2 years ago

Biological evolution occurs on the backs of millions of deaths.

_benj2• 2 years ago

This comment reminded me of the TV show I've been recently watching on Netflix, Pantheon. It's about a different technical breakthrough (I don't want to put any spoilers), but it's also something that completely alters society, no security is able to deal with that new technology, first thing that happens is that the technology is weaponized... etc.

Idk enough about quantum computing to even understand this... but a technology that turns, say, AES or Blowfish, suddenly trivial to crack would very likely change the world

kridsdale16• 2 years ago

This is how I feel about drones.

seanw444• 2 years ago

They're not. What's there to think about?

preisschild• 2 years ago

As if "thinking about it" will ever stop people from acting first.

I'm far more scared when tech-bros like Musk land on Mars and contaminate stuff we might not even be able to detect yet.

whimsicalism8• 2 years ago

i'm not even remotely 'far more scared' about that. i think you are insufficiently scared about crypto being broken

taf22• 2 years ago

Is this using ionq or is this in-house from google?

vessenes7• 2 years ago

They say in-house with their own US fab in the announcement.

DebtDeflation2• 2 years ago

the more qubits we use in Willow, the more we reduce errors, and the more quantum the system becomes

That's an EXTRAORDINARY claim and one that contradicts the experience of pretty much all other research and development in quantum error correction over the course of the history of quantum computing.

wasabi99101124• 2 years ago

It's really not so extraordinary, exponential reduction in logical errors when the physical error rate is below a threshold (for certain types of error correcting codes_ is well accepted an both theoretical and computational grounds.

For a rough but well-sourced overview, see Wikipedia: https://en.wikipedia.org/wiki/Threshold_theorem

For a review paper on surface codes, see A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, “Surface codes: Towards practical large-scale quantum computation,” Phys. Rev. A, vol. 86, no. 3, p. 032324, Sep. 2012, doi: 10.1103/PhysRevA.86.032324.

DebtDeflation2• 2 years ago

Does this not assume uncorrelated errors?

qnleigh10• 2 years ago

I wouldn't call it extraordinary, as this has been expected since the first quantum error correcting codes were worked out theoretically. But it is a strong claim, backed up with comparably strong evidence. Figure 1d of the paper shows exactly this https://arxiv.org/html/2408.13687v1, and unlike many other comparable works, there are no hat tricks like post-selection to boost the numbers.

deepburner2• 2 years ago

it... doesn't? threshold theorems are well known.

da-bacon• 2 years ago

That's an EXTRAORDINARY claim and one that contradicts the experience of pretty much all other research and development in quantum error correction over the course of the history of quantum computing.

Not sure why you would say that? This sort of exponential suppression of errors is exactly how quantum error correction works and why we think quantum computing is viable. Source: have worked on quantum error correction for a couple of decades. Disclosure: I work on the team that did this experiment. More reading: lecture notes from back in the day explaining this exponential suppression https://courses.cs.washington.edu/courses/cse599d/06wi/lectu...

pjs_• 2 years ago

It's a categorically new experimental regime but it's exactly what was supposed to happen. I think it's an awesome result.

deanCommie3• 2 years ago

I don't want to judge people by their cover, but I want to confess to having those feelings right now.

In this day and age, I feel an immediate sense of distrust to any technologist with the "Burning Man" aesthetic for lack of a better word. (which you can see in the author's wikipedia profile from an adjacent festival -> https://en.wikipedia.org/wiki/Hartmut_Neven, as well as in this blog itself with his wristbands and sunglasses -> https://youtu.be/l_KrC1mzd0g?si=HQdB3NSsLBPTSv-B&t=39)

In the 2000's, any embracement of alternative culture was a breath of fresh air for technologists - it showed they cared about the human element of society as much as the mathematics.

But nowadays, especially in a post-truthiness, post-COVID world, it comes off in a different way to me. Our world is now filled with quasi-scientific cults. From flat earthers to anti-vaxxers, to people focused on "healing crystals", to the resurgence of astrology.

I wouldn't be saying this about anyone in a more shall we say "classical" domain. As a technologist, your claims are pretty easily verifiable and testable, even on fuzzy areas like large language models.

But in the Quantum world? I immediately start to approach the author of this with distrust:

* He's writing about multiverses

* He's claiming a quantum performance for something that would take a classical computer septillions of years.

I'm a layman in this domain. If these were true, should they be front page news on CNN and the BBC? Or is this just how technology breakthroughs start (after all the Transformer paper wasn't)

But no matter what I just can't help but feel like the author's choices harm the credibility of the work. Before you downvote me, consider replying instead. I'm not defending feeling this way. I'm just explaining what I feel and why.

bubblyworld• 2 years ago

I don't share your mistrust of the aesthetic, but I think it's pretty natural to be skeptical of the out-group, so to speak, doubly so if you have no practical way of verifying their claims. At least you're honest about it!

I guess something to think about it that amongst a group like the "burners" there is huge variety in individual experience and skill. And even within a single human mind it's possible to have radically groundbreaking thoughts in one domain, and simultaneously be a total crack-pot in another. Linus Pauling and the vitamin C thing comes to mind. There's no such thing as an average person!

I guess we'll see what the quantum experts have to say about this in the weeks to come =)

leptons2• 2 years ago

I know Hartmut Neven personally and professionally, and have for decades. He's not anything like you claim he is. Attacking him for wearing a wristband? That's an ad hominem attack, and not worthy of my time to counter you on.

The fact is that "Burners" are everywhere, nothing about Burning Man means someone is automatically a quack. Your distrust seems misplaced and colored by your own personal biases. The list of prominent people in tech that are also "burners" would likely shock you. I doubt you've ever been to Burning Man, but you're going to judge people who have? Maybe you're just feeling a little bit too "square" and are threatened by people who live differently than you do.

Yes, Hartmut has a style, yes, he enjoys his lifestyle, no, he's not a quack. You don't have to believe me, and I don't expect that you will, but I've talked at length with him about his work, and about a great many other topics, and he is not as you think he is.

Your comment here says far more about you than it says about Hartmut Neven.

deanCommie3• 2 years ago

Attacking him for wearing a wristband? That's an ad hominem attack, and not worthy of my time to counter you on.

I picked my words very carefully and I would appreciate if you responded to what I said, not what you think I implied.

I specifically called out - I'm having feelings of bias. That in a field full of quack science and overpromises and underdelivery, I am extraordinarily suspicious of anyone who I feel might be associated with a shall we say "less than rigorous relationship with scientific accuracy". This person's aesthetic reminds me of this.

The fact is that "Burners" are everywhere, nothing about Burning Man means someone is automatically a quack. Your distrust seems misplaced and colored by your own personal biases. The list of prominent people in tech that are also "burners" would likely shock you. I doubt you've ever been to Burning Man, but you're going to judge people who have? Maybe you're just feeling a little bit too "square" and are threatened by people who live differently than you do.

You couldn't be more wrong. I'm a repeat Burner throughout the 2000's (though it's been a decade), and I've been to a dozen regional Burner events. I know many Burners both in the tech industry and outside of it.

So I actually speak with some experience. I know wonderful people who are purely artists and are not scientifically/technologically inclined - and they're great. I also know deep technologists for whom Burning man is purely an aesthetic preference - a costume not an outfit. Something to pretend to be for a little while but that otherwise has no bearing on their outside life.

And I unfortunately know those whose brainrot ends up intertwining. Crypto evangelists who find healing crystals just as groundbreaking as the blockchain. It's this latter category that I am the most suspicious of, and what I worry when I see a person presented as an authoritative leader in the Quantum Computing domain demonstrate in their external presentation.

I led with an acknowledgement that I am judging a book by it's cover, which one ought to never do. But I think it is worth pointing out because respectability in a cutting edge field is important, lest you end up achieving technological breakthroughs that don't actually change society at all (as already happened with Google Glass).

You don't have to believe me, and I don't expect that you will,

Why would you expect that I wouldn't?

but I've talked at length with him about his work, and about a great many other topics, and he is not as you think he is.

That's fantastic to hear! You have direct evidence contradicting the assumptions generated by my first impression. This is all that matters, and all you had to say.

wasabi99101124• 2 years ago

I know Hartmut Neven personally and professionally, and have for decades

I don't want to put you on the spot too much, but can you speak to why he included the part about many-worlds in this blog post?

I don't know enough about Google to say if maybe someone else less technical wrote that, or if he is being pressured to put sci-fi sounding terms in his posts, or if he believes Google's quantum computer is actually testing many-worlds, or some other reason I can't think of.

wasabi99101124• 2 years ago

I immediately start to approach the author of this with distrust:
* He's writing about multiverses
* He's claiming a quantum performance for something that would take a classical computer septillions of years.
I'm a layman in this domain

I think your skepticism is well-founded. But as you learn more about the field, you learn what parts are marketing/hype bullshit, and what parts are not, and how to translate from the bullshit to the underlying facts.

IMO:

He's writing about multiverses

The author's pet theory, no relevance to the actual science being done.

* He's claiming a quantum performance for something that would take a classical computer septillions of years.

The classical computer is running a very naive algorithm, basically brute-force. It is very easy to write a classical algorithm which is very slow. But still, in the field, it takes new state-of-the-art classical algorithms run on medium size clusters to get results that are on-par with recent quantum computers. Not even much better, just on-par.

Or is this just how technology breakthroughs start (after all the Transformer paper wasn't)

You could say that. It's not truly a breakthrough, but it is one more medium-size step in a rapidly advancing field.

hello_computer2• 2 years ago

The hall of great scientists is packed with holders of strange beliefs. Half of Newton's writings were on religious speculation, alchemy, and the occult. One of Einstein's very favorite books was Blavatsky's "Isis Unveiled". Just about every key person in early QM was deep into the Vedas. Kary Mullis was an AIDS denialist, and questioned the utility of his own test as a virus detector. If you really think about it, you will see that this phenomenon arises more from necessity than coincidence.

Aeium• 2 years ago

The quantum performance thing is real, but that the random circuit sampling problem they are tabling as the benchmark here is for a quantum circuit.

So really what is being claimed is that classical computers can't easily simulate quantum ones. But is that really surprising?

What would be surprising would be that kind of speedup vs classical on some kind of general optimization algorithm. I don't think that is what they are claiming though, even if it does kind of seem like it's being presented that way.

pandemic_region• 2 years ago

Willow performed a standard benchmark computation in under five minutes that would take one of today’s fastest supercomputers 10 septillion years — a number that vastly exceeds the age of the Universe.

What computation would that be?

Also, what is the relationship, if any, between quantum computing and AI? Are these technologies complementary?

ra7• 2 years ago

It's in the article. Random circuit sampling benchmark: https://research.google/blog/validating-random-circuit-sampl...

crote2• 2 years ago

Is it really fair to call that "computation"? I am definitely not an expert, but it seems they are just doing a meaningless operation which happens to be trivial on a quantum computer but near-impossible to simulate on a classical computer.

To me that sounds a bit like saying my "sand computer" (hourglass) is way faster than a classical computer, because it'd take a classical computer trillions of years to exactly simulate the final position of every individual grain of sand.

Sure, it proves that your quantum computer is actually a genuine quantum computer, but it's not going to be topping the LINPACK charts or factoring large semiprimes any time soon, is it?

0xB31B1B• 2 years ago

"Also, what is the relationship, if any, between quantum computing and AI? Are these technologies complementary?"

AI is limited in part by the computation available at training and runtime. If your computer is 10^X times faster, then your model is also "better". Thats why we have giant warehouses full of H100 chips pulling down a few megawatts from the grid right now. Quantum computing could theoretically allow your phone to do that.

spencerflem• 2 years ago

A quantum computer is not just a 10^X faster normal computer.

Are there AI algorithms that would benefit from quantum?

kridsdale16• 2 years ago

Makes sense. My brain is able to do that work on milliwatts.

wasabi99101124• 2 years ago

Also, what is the relationship, if any, between quantum computing and AI? Are these technologies complementary?

Ongoing research.

The main idea of quantum machine learning is that qubits make an exponentially high-dimensional space with linear resources, so can store and compute a lot of data easily.

However, getting the data in and results out of the quantum computer is tricky, and if you need many iterations in your optimization, that may destroy any advantage you have from using quantum computers.

oldgradstudent• 2 years ago

Also, what is the relationship, if any, between quantum computing and AI? Are these technologies complementary?

AI is quite good in producing the meaningless drivel needed for quantum computing related press releases.

sys32768• 2 years ago

In other words, get off the cloud so nobody has your encrypted data which they will be able to crack in a few minutes five or ten years from now?

noident• 2 years ago

It depends on the algorithm you use to encrypt your data.

Only asymmetric cryptography is threatened. There is no realistic threat to symmetric encryption like AES.

If you are encrypting your cloud data with ed25519 or RSA, then yes, a quantum computer could theoretically someday crack them.

Havoc2• 2 years ago

Only asymmetric cryptography is threatened.

aka everything we use daily

codeulike2• 2 years ago

They opened the API for it and I'm sending requests but the response always comes back 300ms before I send the request, is there a way of handling that with try{} predestined{} blocks? Or do I need to use the Bootstrap Paradox library?

timcobb• 2 years ago

What does Gemini say?

KTibow• 2 years ago

It responds with 4500 characters: https://hst.sh/olahososos.md

wk_end• 2 years ago

Finally, INTERCAL’s COME FROM statement has a practical use.

handfuloflight• 2 years ago

Have you tried using the Schrödinger Exception Handler? It catches errors both before and after they occur simultaneously, until you observe the stack trace.

oblio• 2 years ago

I swear I can't tell which of these comments are sarcastic/parody and which are actual answers.

A sort of quantum commenting conundrum, I guess.

nobrains• 2 years ago

What happens when you don't send the request after receiving the response? Please try and report back.

r3trohack3r• 2 years ago

You unlock the "You've met a terrible fate." achievement [1]

[1] https://outerwilds.fandom.com/wiki/Achievements

ukuina• 2 years ago

Please report back and try.*

tealpod• 2 years ago

Looks like we don't have a choice.

jmcqk6• 2 years ago

No matter what we've tried so far, the request ends up being sent.

The first time I was just watching, determined not to press the button, but when I received the response, I was startled into pressing it.

The second time, I just stepped back from my keyboard, and my cat came flying out of the back room and walked on the keyboard, triggering the request.

The third time, I was holding my cat, and a train rumbled by outside, rattling my desk and apparently triggering the switch to send the request.

The fourth time, I checked the tracks, was holding my cat, and stepped back from my keyboard. Next thing I heard was a POP from my ceiling, and the request was triggered. There was a small hole burned through my keyboard when I examined it. Best I can figure, what was left of a meteorite managed to hit at exactly the right time.

I'm not going to try for a fifth time.

cloudking2• 2 years ago

The answer is yes and no, simultaneously

yu3zhou4• 2 years ago

Did you try staring on your IP packets while sending the requests?

Nifty3929• 2 years ago

Try using inverse promises. You get back the result you wanted, but if you don't then send the request the response is useless.

It's a bit like Jeopardy, really.

dtquad2• 2 years ago

They opened the API for it and I'm sending requests but the response always comes back 300ms before I send the request

For a brief moment I thought this was some quantum-magical side effect you were describing and not some API error.

pinkmuffinere• 2 years ago

Isn't that.... the joke?

vangamoZX• 2 years ago

Write the catch clause before the try block

sebastiennight2• 2 years ago

You are getting that response 300ms beforehand because your request is denied.

If you auth with the bearer token "And There Are No Friends At Dusk." then the API will call you and tell you which request you wanted to send.

GuB-42• 2 years ago

It think you are supposed to use a "past" object to get your results before calling the API.

tiborsaas• 2 years ago

Try setting up a beam-splitter router and report back with the interference pattern. If you don't see a wave pattern it might be because someone is spying on you.

yonatan8070• 2 years ago

Pretty sure you just need to use await-async (as opposed to async-await)

codeulike2• 2 years ago

Update: I tried installing the current Boostrap Paradox library but it says I have to uninstall next years version first.

nick3443• 2 years ago

Help! Every time I receive the response, an equal number of bits elsewhere in memory are reported as corrupt by my ECC RAM.

robomartin• 2 years ago

I'm sending requests but the response always comes back 300ms before I send the request

Ah. Newbie mistake. You need to turn OFF your computer and disconnect from the network BEFORE sending the request. Without this step you will always receive a response before the request is issued.

qingcharles• 2 years ago

I'm trying to write a new version of Snake game in Microsoft Q# but it keeps eating its own tail.

kernal• 2 years ago

Willow’s performance on this benchmark is astonishing: It performed a computation in under five minutes that would take one of today’s fastest supercomputers 1025 or 10 septillion years. If you want to write it out, it’s 10,000,000,000,000,000,000,000,000 years. This mind-boggling number exceeds known timescales in physics and vastly exceeds the age of the universe. It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch.

A much simpler explanation is that your benchmark is severely flawed.

wasabi99101124• 2 years ago

"Severely flawed" is a matter of interpretation, and I don't want to argue for or against.

But to put into context, these numbers are likely accurate, but represent the time it would take for a very naive classical algorithm (possibly brute-force, I am unsure).

For example, the previous result claimed it would take Summit 10,000 years to do the same calculation as the Sycamore quantum chip. However, other researchers were able to reproduce results classically using tensor-network-based methods in 14.5 days using a "relatively small cluster". [1]

[1] G. Kalachev, P. Panteleev, P. Zhou, and M.-H. Yung, “Classical sampling of random quantum circuits with bounded fidelity,” arXiv.org, https://arxiv.org/abs/2112.15083 (accessed Dec. 9, 2024).

xnx• 2 years ago

Is anyone else even close to Google in this space? (e.g. on the "System Metrics" the blog defines)

wasabi99101124• 2 years ago

I would expect IBM, but I can't find any information on their system metrics based on a quick google search.

Would love if someone could weight in.

lanthissa• 2 years ago

not publicly

taf22• 2 years ago

IonQ - they are powering AWS solution here: https://aws.amazon.com/braket/quantum-computers/ionq/

Not sure if they are close in terms of specs but looks like they are a viable solution and seeing an increase in utilization over the last year... Seems both are pretty interesting to keep an eye on.

EvgeniyZh2• 2 years ago

Main players IMHO are IBM and Quantinuum, the latter employing different platform (ions). Neither could perform the same experiment I think, but have their own advantages. QuEra also looks good but are not as mature yet imho.

paxys2• 2 years ago

Give it a few more years and a smaller, more focused company will come in and launch a successful product based on their research.

cahaya• 2 years ago

I'm wondering how this Google announcement will have an impact on their future revenue, and how easily competitors can replicate this breakthrough?

krick• 2 years ago

I would be very grateful if somebody can point to an actually good blogpost/video that would summarize the current state of the domain. I mean, I remember some "quantum annealing" providers as far as some 10 years ago (I mean, as a service, as such D-Wave exists for 25 years now), but I never actually learned if they are truly useful for anything (like, real numbers, what amount of computation these thing can perform and if it's truly cheaper/faster than throwing a bunch of GPUs on it). From time to time there are some news feturing dope photos, about some new chip from IBM, that is useful for nothing, but a big breakthrough for reasons I don't understand.

But I don't really have a feel of what's going on, really. How many quantum computers there are, is there anything that is actually capable of performing anything more than just being an ongoing research prototype? Some educated guesses about how far can be some non-public projects by now? Like, is it possible that some secret CIA project is further ahead than what we know, or if it's even more unlikely and farther away than fusion power? Or maybe it's even more comparable to cold fusion?

I know, that this kinda exists as an idea, and apparently somebody's working on it, but that's pretty much it.

fguerraz• 2 years ago

Am I oversimplifying in thinking that they’ve demonstrated that their quantum computer is better than at simulating a quantum system than a classical computer?

In which case, should I be impressed? I mean sure, it sounds like you’ve implemented a quantum VM.

TachyonicBytes2• 2 years ago

Simulating a quantum system is a hard challenge and it's actually how Feynman proposed the quantum computing paradigm in the first place. It's basically the original motive.

freetonik• 2 years ago

Exactly.

I’ve seen lots of people dismissing this as if it isn’t impressive or important. I’ve watched one video where the author said in a deprecating manner “quantum computers are good for just two things: generating random numbers and simulating quantum systems”.

It’s like saying “the thing is good for just two things: making funny noises and producing infinite energy”.

(Also, generating random numbers is pretty useful, but I digress)

slater• 2 years ago

I bet Vimeo videos will still chug on it

TachyonicBytes2• 2 years ago

Link to the actual article: https://www.nature.com/articles/s41586-024-08449-y

qnleigh10• 2 years ago

and the arXiv preprint, which isn't paywalled https://arxiv.org/html/2408.13687v1

wasabi99101124• 2 years ago

And contains more information, figures, and lots of supplementary material.

zelon88• 2 years ago

It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch.

Can someone explain to me how he made the jump from "we achieved a meaninful threshold in quantum computing performance" to "The multiverse is probably real."

dekhn• 2 years ago

The explanation is that Hartmut Neven has a bunch of sci-fi beliefs and somehow has managed to hold onto his job and even get to write parts of press releases.

not_a_bot_4sho• 2 years ago

My money is on edibles.

radioactivist• 2 years ago

Some of these results have been on the arxiv for a few months (https://arxiv.org/abs/2408.13687) -- are there any details on new stuff besides this blog post? I can't find anything on the random circuit sampling in the preprint (or its early access published version).

qnleigh10• 2 years ago

The peer-reviewed version at Nature has more technical details about the processor itself.

ryandvm• 2 years ago

Imagine your civilization develops quantum computing technology and it's for... advertising.

"What is their mission? Cure cancer? Eliminate poverty? Explore the universe? No, their goal: to sell another fucking Nissan." --Scott Galloway

mperham2• 2 years ago

That's how you monetize attention, digital consumption. If you aren't paying for it, you are the product being sold.

NoOn3• 2 years ago

In modern world, sometimes, even if you pay for it, it doesn't always give guarantees... :(

softwaredoug2• 2 years ago

With this plus the weather model announcement. I’m curious what people think about the meta question on why corporate labs like Google DeepMind etc seem to make more progress on big problems than academia?

There are a lot of critiques about academia. In particular that it’s so grant obsessed you have to stay focused on your next grant all the time. This environment doesnt seem to reward solving big problems but paper production to prove the last grant did something. Yet ostensibly we fund fundamental public research precisely for fundamental changes. The reality seems to be the traditional funding model create incremental progress within existing paradigms.

LeftHandPath5• 2 years ago

I did quantum computing research in university. We did meaningful work and published meaningful research.

Around 50% of our time was spent working in Overleaf making small improvements to old projects so that we could submit to some new journal or call-for-papers. We were always doing peer review or getting peer reviewed. We were working with a lot of 3rd-party tools (e.g. FPGAs, IBM Q, etc). And our team was constantly churning due to people getting their degrees and leaving, people getting too busy with coursework, and people deciding they just weren't interested anymore.

Compare that to the corporate labs: They have a fully proprietary ecosystem. The people who developed that ecosystem are often the ones doing research on/with it. They aren't taking time off of their ideas to handle peer-review processes. They aren't taking time off to handle unrelated coursework. Their researchers don't graduate and start looking for professor positions at other universities.

It's not surprising in the slightest that the corporate labs do better. They're more focused and better suited for long-term research.

softwaredoug2• 2 years ago

I wonder what makes research different than product development at a company?

Because in product development, there can be short-sighted industry decisions based on quarterly returns. I've also seen a constant need to justify outcomes based on KPIs etc, and constantly justifying your work, etc.

anonymousDan• 2 years ago

Where exactly do you think the idea for a quantum computer came from in the first place?

lysecret• 2 years ago

I can only speak for the weather models (since it is in my domain). The answer is that the issues are much more engineering and scaling/infra issues than theoretical issues, and google is good at engineering (or attracts people that are).

andrewla2• 2 years ago

This is yet another attempt to posit NISQ results (Noisy Intermediate Scale Quantum) as demonstrations of quantum supremacy. This does not allow us to do useful computational work; it's just making the claim that a bathtub full of water can do fluid dynamic simulations faster than a computer with a bathtub-full-of-water-number-of-cores can do the same computation.

If history is any guide we'll soon see that there are problems with the fidelity (the system they use to verify that the results are "correct") or problems with the difficulty of the underlying problem, as happened with Google's previous attempt to demonstrate quantum supremacy [1].

[1] https://gilkalai.wordpress.com/2024/12/09/the-case-against-g... -- note that although coincidentally published the same day as this announcement, this is talking about Google's previous results, not Willow.

cryptozeus2• 2 years ago

QTUM

cloudking2• 2 years ago

Contains MSTR and not GOOG...

cryptozeus2• 2 years ago

Point is to get in before goog gets in

djoldman• 2 years ago

I wonder if anyone else will be forced to wait on https://scottaaronson.blog/ to tell us if this is significant.

gloriousduke• 2 years ago

I was about to add a similar comment. Definitely interested to read his evaluation and whether there is more hype than substance here, though I'm guessing it may take some time.

EvgeniyZh2• 2 years ago

He told when the preprint was published

https://scottaaronson.blog/?p=8310

qnleigh10• 2 years ago

He's already blogged about it a bit here

https://scottaaronson.blog/?p=8310#comments

and here

https://scottaaronson.blog/?p=8329

though I bet he will have more to say now that the paper is officially out.

arresin• 2 years ago

This is weird. I got this pop up halfway through reading:

After reading this article, how has your perception of Google changed? Gotten better Gotten worse Stayed the same

beyondCritics• 2 years ago

It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch.

Makes sense, or doesn't it? What's your take on the multiverse theory?

dom96• 2 years ago

Can anyone comment on how this chip is built? What does the hardware look like?

stan_kirdey• 2 years ago

Is anyone fine-tuning llama to write Q#? I feel LLMs can be a helpful tool in learning how to code quantum systems.

dtquad2• 2 years ago

Do Americans still want to breakup the big US tech companies like Google? With proper regulation it feels like their positive externalities, like this, is good for humanity.

0_____0• 2 years ago

"Proper regulation" may involve breaking companies into pieces such that they cannot dominate industries and deprive the public the option of choosing a different provider for the services they provide. Does an Alphabet subsidiary working on quantum computer research require 90% of search traffic to go through Google? Or for Android handsets to send a really phenomenal amount of telemetry back to the mothership with no real recourse for the average user?

skort• 2 years ago

The key words here are "proper regulation". In an era where industries have captured governmental bodies, there will likely be no such regulation, and these tech companies will continue to siphon up resources and funnel them to a handful at the top.

A quote from the article is especially ludicrous: > to benefit society by advancing scientific discovery, developing helpful applications, and tackling some of society's greatest challenges

You don't need a quantum computer to do this. We can solve housing and food scarcity today, arguably our greatest challenges. Big tech has been claiming that it's going to solve all of our problems for decades now and it has yet to put up.

If you want this type of technology to be made and do actual good, we need publicly funded research institutions. Tech won't save us.

noisy_boy• 2 years ago

Regulation is the handle of the chainsaw aka corporation. Right now it is a bit wobbly which explains the cuts on the society we see around us. The faster we make the blades turn due to technological advances, the firmer the handle needs to be.

correlator• 2 years ago

I met julian touring UCSB as perspective grad students. We sat together at dinner and he was really smart, kind, and outgoing. Great to see him presenting this work!

robot• 2 years ago

there is so much skepticism on quantum computing that instead of inflated marketing words one should always start by what the biggest problems are, how they are not still solved yet, and then introduce what the new improvement is.

Otherwise there is no knowing if the accomplishment is really significant or not.

gauge_field2• 2 years ago

Especially if you consider how they choose the words and how they can be interpreted

" Second, Willow performed a standard benchmark computation in under five minutes that would take one of today’s "

Standard benchmark in what sense. Well, it was chosen for task where quantum computer would have better performance.

I am not saying this is nothing. Maybe, use more reserved words, e.g. "special quantum oriented benchmark" or something.

When I think of standard benchmark, I am thinking more common scenarios, e.g. searching, sorting, matrix multiplication.

staunton• 2 years ago

This is a great technical achievement. It gives me some hope to see that the various companies are able to invest into what is still very basic science, even if it were mostly as vanity projects for advertising purposes.

Quantum computing will surely have amazing applications that we cannot even conceive of right now. The earliest and maybe most useful applications might be in material science and medicine.

I'm somewhat disappointed that most discussions here focus on cryptography or even cryptocurrencies. People will just switch to post-quantum algorithms and most likely still have decades left to do so. Almost all data we have isn't important enough that intercept-now-decrypt-later really matters, and if you think you have such data, switch now...

Breaking cryptography is the most boring and useless application (among actual applications) of quantum computing. It's purely adversarial, merely an inconsequential step in a pointless arms race that we'd love to stop, if only we could learn to trust each other. To focus on this really betrays a lack of imagination.

rgmerk• 2 years ago

“Quantum computing will surely have amazing applications that we cannot even conceive of right now…”

As best I understand, it’s not clear yet whether quantum computing will ever have any practical applications.

Furthermore, there has already been a great deal of work identifying potential applications for a quantum computer, so I’d say we’ve got a fair idea of what you could do with one if it ever exists.

gauge_field2• 2 years ago

The first (and important enough) practical application (apart from breaking RSA) seems to be simulation of quantum systems. But, this has some problems since alot of quantum models can be simulated with Deep Learning methods (getting better) to a good approximation. Probably, some quantum systems wont be good for (classical) Deep Learning, those might be the first candidates for application of quantum computers.

hi412• 2 years ago

What benchmark is being referred here?

>Willow performed a standard benchmark computation in under five minutes that would take one of today’s fastest supercomputers 10 septillion (that is, 1025) years — a number that vastly exceeds the age of the Universe

dbrans• 2 years ago

https://research.google/blog/validating-random-circuit-sampl...

hi412• 2 years ago

If error is getting corrected, doesn’t it mean lower entropy? If so where else is entropy increasing, if it is a valid question to be asked.

roywiggins2• 2 years ago

Probably the same as normal chips: waste heat?

codeyperson• 2 years ago

If we are in a simulation. This seems like a good path to getting our process terminated for consuming too much compute.

prettyStandard3• 2 years ago

AI too.

I've followed Many worlds & Simulation theory a bit too far and I ended up back where I started.

I feel like the most likely scenario is we are in a AI (kinder)garden being grown for future purposes.

So God is real, heaven is real, and your intentions matter.

Obviously I have no proof...

swores2• 2 years ago

"and your intentions matter."

How do you reach that conclusion?

Characters in The Sims games technically have us human players as gods, it doesn't mean that when we uninstall the game those characters get to come into our earthly (to them) heaven or have any consequences for actions performed during the simulation?

a2128• 2 years ago

I doubt we would even register as a blip. The universe is absolutely massive and there's celestial events that are unthinkably massive and complex. Black hole mergers, supernovae, galaxies merging. Hell, think of what chaos happens in the inside of our own sun, and multiply that by 100 billion stars in a galaxy, and multiply that by 100 billion galaxies. Humanity is ultimately inconsequential.

swores2• 2 years ago

Surely it would depend on what the simulation actually was?

If you imagine simulations we can build ourselves, such as video games, it's not hard to add something at the edge of the map that users are prevented from reaching and have the code send "this thing is massive and powerful" data to the players. Who's to say that the simulation isn't actually focussed on earth, and everything including the sun is actually just a fiction designed to fool us?

rkagerer• 2 years ago

How much support infrastructure does this thing need? (eg. Cryogenic cooling?) How big is a whole 'computer' and how much power draw?

Havoc2• 2 years ago

So is it time to 100x key length on browsers yet or not?

bwesterb4• 2 years ago

About 8x just for key agreement and 40x for signatures. It's a lot. For key agreement it's worth it, and now about 1/3 of browsers in the wild use it. http://radar.cloudflare.com/adoption-and-u…

jxdxbx• 2 years ago

More misleading random circuit sampling benchmarks. All it proves is that Google has built a quantum computer that does quantum things.

webdevver• 2 years ago

what is this actually useful for?

mvkel• 2 years ago

Seems to coincide with Google's annual quantum announcement, which has happened each fall since 2014

riku_iki• 2 years ago

annual perf cycle probably

rvz• 2 years ago

I don’t usually say this often, unlike most here but this is actually a huge achievement in quantum computing.

Yet another example as to why Google is essentially not going anywhere or ‘dying’ as most have been proclaiming these days.

Uptrenda• 2 years ago

'It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch.'

Wait... what? Google said this and not some fringe crackpot?

GilKalai• 2 years ago

In the past five years I participated in a project (with Yosi Rinott and Tomer Shoham) to carefully examine the Google's 2019 "supremacy" claim. A short introduction to our work is described here: https://gilkalai.wordpress.com/2024/12/09/the-case-against-g.... We found in that experiment statistically unreasonable predictions (predictions that were "too good to be true") indicating methodological flaws. We also found evidence of undocumented global optimization in the calibration process.

In view of these and other findings my conclusion is that Google Quantum AI’s claims (including published ones) should be approached with caution, particularly those of an extraordinary nature. These claims may stem from significant methodological errors and, as such, may reflect the researchers’ expectations more than objective scientific reality.

janpmz• 2 years ago

I've heared claims before that quantum computers are not real. But I didn't understand it. Can anybody explain the reasoning behind the criticism? Are they just simulated?

Rhapso• 2 years ago

So the basics:

- quantum physics are real, this isn't about debating that. The theory underpinning quantum computing is real.

- quantum annealing is theoretically real, but not the same "breakthrough" that a quantum computer would be. Z-wave and google have made these.

- All benchmark computations have been about simulating a smaller quantum computer or annealer. which these systems can do faster than a brute force classical search. These are literally the only situation where "quantum supremacy" exists.

- There is literally no claim of "productive" computation being made by a quantum computer. Only simulations of our assumptions about quantum systems.

- The critical gap is "quantum error correction", proof that they can use many error prone physical qubits to simulate a smaller system with lower error. There isn't proof yet that is actually possible.

This result they are claiming, is they have "critical error correction" is the single most groundbreaking result we could have in quantum computing. Their evidence does not satisfy the burden of proof. They also only claim to have 1 qubit, which is intrinsically useless, and doesn't examine the costs of simulating multiple interacting qubits.

sgt101• 2 years ago

I think that this is now a very fringe position - the accumulation of reported evidence is such that if folks were/are fooling themselves it would now be a case of an entire community conspiring to keep things quiet. I mean, it's not impossible that could happen, but I think it would be extraordinary.

On the other hand the question is what does "real QC" mean? The current QC's perform very limited and small computations, they lack things like quantum memory. The large versions are extremely impractical to use in the sense that they run for 1000ths of a second and take many hours to setup for a run. But that doesn't mean that the physical effects that they use/capture aren't real.

Just a long long way from practical.

whoitwas2• 2 years ago

I'm skeptical only because they have AI in the name.

qnleigh10• 2 years ago

What experimental data would you need to see to change your mind? Either from Google or from another player?

attentionmech2• 2 years ago

What does it mean when they say that the computations are happening in multiverse? I didn't know we are that advanced already :)

attentionmech2• 2 years ago

I don't know why i got downvoted. "It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch." -- this is from the article itself.

raminf• 2 years ago

Relatives were asking for a basic explainer. Here's a good one by Hannah Fry: https://youtu.be/1_gJp2uAjO0

crishoj• 2 years ago

Take the announcement with a grain of salt. From German physicist Sabine Hoffenfelder:

The particular calculation in question is to produce a random distribution. The result of this calculation has no practical use. They use this particular problem because it has been formally proven (with some technical caveats) that the calculation is difficult to do on a conventional computer (because it uses a lot of entanglement). That also allows them to say things like "this would have taken a septillion years on a conventional computer" etc. It's exactly the same calculation that they did in 2019 on a ca 50 qubit chip. In case you didn't follow that, Google's 2019 quantum supremacy claim was questioned by IBM pretty much as soon as the claim was made and a few years later a group said they did it on a conventional computer in a similar time.

https://x.com/skdh/status/1866352680899104960

l33tman2• 2 years ago

TBH you need to take the youtube influencer Sabine Hoffenfelder with a bigger grain of salt. She has converted to mainly posting clickbait youtube stuff over the last years (unfortunately, she was interesting to listen to earlier).

The RCS is a common benchmark with no practical value, as is stated several times in the blog announcement as well. It's used because if a quantum computer can't do that, it can't do any other calculation either.

The main contribution here seems to be what they indeed put first, which is the error correction scaling.

falleng0d• 2 years ago

I have observed the change in approach to use aggressive (clickbaity) thumbnails, but I do think the quality of the content has not changed.

And I can't blame her for adopting this trend, in many cases it is the difference between surviving or not on YouTube nowadays.

Closi• 2 years ago

I think simplifying her to 'youtube influencer' is unfair - she is a doctor of theoretical physics with a specialism in quantum gravity who produces science content for youtube. She knows the field enough to comment.

She doesn't even say that this isn't a big leap (she says it's very impressive - just not the sort of leap that means that there are now practical applications for quantum computers, and that a pinch of salt is required on the claim of comparisons to a conventional computer due to the 2019 paper with a similar benchmark).

pixelsort• 2 years ago

As a counterpoint, she recently reviewed a paper in one of her recent videos and completely tore it to shreds as apparently the math was full of absolute nonsense.

This was a fascinating watch, and not the kind of content that is easy to find. Besides videos like that one, I enjoy her videos as fun way to absorb critical takes on interesting science news.

Maybe she is controversial for being active and opinionated on social media, but we need more science influencers and educators like her, who don't just repeat the news without offering us context and interpretation.

zipy124• 2 years ago

Just because she is a YouTuber doesn't diminish her other credentials, just as she is incetivised to do clickbait, so are actual scientific communication outlets such as nature, and the more clicky they are the more downloads and citation they will acquire. Incentives change content but don't directly detract from someone's expertise. See: the fact that most universities now publish some lectures on YouTube, it doesn't make the content any less true.

cowl• 2 years ago

she was right the first time when they announced this in 2019 and this time even they admit in their own press release:

Of course, as happened after we announced the first beyond-classical computation in 2019, we expect classical computers to keep improving on this benchmark

As IBM showed their estimate of classical computer time is taken out of their a**es.

perching_aix• 2 years ago

They explicitly cover all of these caveats in the announcement.

Problems that benefit from quantum computing as far as I'm aware have their own formal language class, so it's also not like you have to consider Sabine's or any other person's thoughts and feelings on the subject - it is formally demonstrated that such problems exist.

Whether the real world applications arrive or not, you can speculate for yourself. You really don't need to borrow the equally unsubstantiated opinion of someone else.

eigenket• 2 years ago

The formal class is called BQP, in analogy with the classical complexity clas BPP. BQP contains BPP but there is no proof that it is stictly bigger (such a proof would imply P != NP). There are problems in BQP we expect are not in BPP but its not clear if there are any useful problems in BQP and not in BPP, other than essentially Shor's algorithm.

On the other hand it's actually not completely necessary to have a superpolynomial quantum advantage in order to have some quantum advantage. A quantum computer running in quadratic time is still (probably) more useful than a classical computer running in O(n^100) time, even though they're both technically polynomial. An example of this is classical algorithms for simulating quantum circuits with bounded error whose runtime is like n^(1/eps) where eps is the error. If you pick eps=0.01 you've got a technically polynomial runtime classical algorithm but it's runtime is gonna be n^100, which is likely very large.

chvid• 2 years ago

It would be interesting to see what "standard benchmark computation" was used and what its implementation would like in a traditional computer language.

Does anyone know?

vctrnk• 2 years ago

Not to defend Google, but they end up saying much the same:

The next challenge for the field is to demonstrate a first "useful, beyond-classical" computation on today's quantum chips that is relevant to a real-world application. We’re optimistic that the Willow generation of chips can help us achieve this goal. So far, there have been two separate types of experiments. On the one hand, we’ve run the RCS benchmark, which measures performance against classical computers but has no known real-world applications. On the other hand, we’ve done scientifically interesting simulations of quantum systems, which have led to new scientific discoveries but are still within the reach of classical computers. Our goal is to do both at the same time — to step into the realm of algorithms that are beyond the reach of classical computers and that are useful for real-world, commercially relevant problems.

_joel• 2 years ago

But can it run Crysis?

whiplash451• 2 years ago

Newbie's question: how far is the RCS benchmark from a more practical challenge such as breaking RSA?

The article concludes by saying that the former does not have practical applications. Why are they not using benchmarks that have some?

Hilift• 2 years ago

There are already companies selling 5,000+ qubit quantum systems to the NSA. I would assume they are already using it for that in the huge data center in Utah.

underdeserver• 2 years ago

Which companies? And how big an RSA key do 5,000+ qubit quantum systems break?

makeworld2• 2 years ago

Source?

wasabi99101124• 2 years ago

Very far.

I'm not sure how to put it quantitatively, but my impression from listening to experts give technical presentations is that the breaking-rsa-type algorithms are a decade or two away.

This is very soon from a security perspective, as all you need is to store current data and break it in the future. But it is not soon enough to use for benchmarking current systems.

ramonverse• 2 years ago

Can't wait for Google to release a breakthrough paper in 5 years just for the authors to leave and build OpenQuant

machina_ex_deus• 2 years ago

You learn a lot by what isn't mentioned. Willow had 101 qubits in the quantum error correction experiment, yet only mere 67 qubits in the random circuit sampling experiment. Why did they not test random circuit sampling with the full set of qubits? Maybe when turning on the full 101 set of qubits, qubits fidelity dropped.

Remember macroscopic objects have 10^23=2^76 particles, so until 76 qubits are reached and exceeded, I remain skeptical that the quantum system actually exploits an exponential Hilbert space, instead of the state being classically encoded by the particles somehow. I bet Google is struggling just at this threshold and they don't announce it.

singlewind• 2 years ago

Now, this is really make other countries nervous. Basically, existing cryptography technology is in danger.

cubefox• 2 years ago

Notice how the most interesting part, the image with their quantum computing roadmap, has too low a resolution to read the relevant text at the bottom. Come on Google.

EGreg• 2 years ago

Any chance people will actually start reversing SHA1 hashes in the next few years? Is there any quantum algorithm for reversing one-way functions like that? (I mention SHA1 because they can find collissions.)

yalogin• 2 years ago

Trying to understand how compute happens in Quantum computers. Is there a basic explanation of how superposition leads to computing?

From chatgpt, "with n qubits a QC can be in a superposition of 2^n different states. This means that QCs can potentially perform computations on an exponential number of inputs at once"

I don't get how the first sentence in that quote leads to the second one. Any pointers to read to understand this?

johndhi• 2 years ago

Genuinely curious: does this make US regulators second-guess breaking up Google? Having a USA company be the first to develop quantum computing would be a major national security advantage.

Willow, Our Quantum Chip

Related Stories