Insert D-Wave Post Here | Hacker News Reader

Insert D-Wave Post Here

scottaaronson.com

119 pointsseycombi52 comments

D-Wave is the original Theranos.

- They make bold claims that many people plainly say don't make sense, before seeing any evidence.

- They use hype to buy time to do scientific research.

- The successes of the D-Wave machine have the same characteristic p-hacking profile of Theranos's Edison machine. That is, try 200 problems, and find 1 that appears to work, which could just be due to chance.

- Apologists talk about "there are a lot of smart people working" there or blame "marketing," but who haven't seen the product and don't know any more than anyone else.

The main differences are (1) the lack of an enigmatic founder. (2) D-Wave publishes, and then in later analysis their claims don't hold up.

Two major differences:

- D-Wave does seem to understand the criticisms and slowly incorporate them into each subsequent test. Theranos instead preferred ad hominem attacks on their attackers.

- The question about D-Wave isn't whether or not it works, it's about whether or not it is useful. Theranos couldn't even get their non-magic technology working right.

The question is how it works, and whether or not it works as advertised. In that sense, no it does not work.

What would explain the interest Google, NASA and the DoD have in D-Wave and the fact that they have acquired a few of these devices?

Most of those organizations are happy to dump $1 million into something with so much potential power on the off-chance that it turns out to work. In fact, if I was in charge of Google, and understood the extremely tiny chance that D-Wave is legit, I would still dump in money, because the tiny chance it works still makes it worth it.

But they (in particular, Google) also have experts in quantum physics and computing on staff, so they're much better equipped than most to determine the validity of the claim.

The actual price for a D-Wave machine is more than $10M. It's a good sign that the first customers have continued their support by upgrading to the newer models.

The DoD will give you a few thousand USD if you write a good enough proposal for the military benefits of sharks with freaking lasers on their heads for the defense of navy vessels [1].

Throwing a few million at a prototype computer on the off chance it works is well within the budget.

[1] Austin Power reference not a real research project.

We do however, since the 1940s, have dolphins with bombs.

As a taxpayer I'm glad those organizations buy these machines, just as they bought Connection Machines, Crays, etc. Some of those bets panned out. And in fact if that money went into nothing but more understanding of the difficulties of quantum computing wouldn't you be glad?

But I don't make the mistake of interpreting a purchase as some sort of endorsement. It's merely an endorsement that "your plausibility level is above our noise floor".

That is way overblown.

What they have is more akin to an analog computer, using quantum tricks to solve problems. The issue here is that most people disagree that this fits into the general definition of a "quantum computer".

That's not what Aaron is saying. He's saying there's no evidence that the D-Wave machine is using any quantum tricks to solve problems at all.

I think this is an unfair comparison (and I have grave doubts of D-wave's "achievements"). People can actually buy the machines and play with them. I have credulous and doubting friends at google who play with/have played with google's D-wave.

Also Theranos explicitly went against the FDA (like 23andMe), while D-wave is going along the publication route.

Theranos was simply bogus (I wouldn't call it fraudulent, but definitely bogus). D-Wave is like many tech startups, some of which do end up panning out.

One of D-wave's founders is now trying to patent teleoperation of robots and pass it off as "novel": http://spectrum.ieee.org/automaton/robotics/artificial-intel...

Also, tossing in some machine learning which I presume they're also trying to patent (because it's not as if UC Berkeley and Google have done that 15 times a year since 2010)....

All this makes me think they're not legit.

Here is MIT Technology Review's article on www.kindred.ai

Kindred AI is teaching robots new tasks using human virtual-reality “pilots.” The ultimate goal is to create a new kind of artificial intelligence.

https://www.technologyreview.com/s/603745/how-a-human-machin...

Here's the list of D-Wave supporters (organizations that have spent real $): Google, NASA, Lockheed Martin, Los Alamos National Labs, University of Southern California, Temporal Defense Systems, US government, CIA/In-Q-Tel. At over $10M per machine, this adds up to at least $60M. Surely the critics in this blog are missing something.

Argument from authority, really?

It seems to be more argument from money spent. Which is equally fallacious, if less likely to show up on a standard list of fallacies.

Wait till you see the list of respected Enron and Theranos supporters...

Do you have information not available to the organizations that have spent over $60M on D-Wave computers? Why are you wasting your time here?

>Why are you wasting your time here?

Because I like reading arguments from authority -- especially ones based on the premise that major organizations never buy into BS.

Which I guess explains why software sold to enterprises is known for its quality and utility.

OK, but that doesn't explain why D-Wave has so many paying customers (Google, NASA, Lockheed Martin, Los Alamos National Labs, University of Southern California, Temporal Defense Systems, US government, CIA/In-Q-Tel).

Even if one insists on "argument from authority", paying customers means nothing.

Satisfied customers, now, that's a whole other thing.

Repeating that list in your response doesn't underscore your point, it instead emphasizes how little you have to add in rebuttal.

Several of the above investors have a well-known history of investing in extreme, improbable avenues of development; whether the idea is that "we've got vast hordes of money—in many cases, other people's money who don't demand much accountability—and if we get lucky one in ten thousand times, that's potentially enough to justify it", or "even if this fails, it might have some side benefits", or, "just by investing in this, we force our adversaries to try spend resources trying to figure if we know something critical that they don't", or some combination, is less clear. But I wouldn't take small investments from those players (and the scale we are talking is small for them) as a sign of probable success.

Yet there are few people alive more qualified to comment on this type of work than Scott Aaronson. Plus, most of his criticism is only directed towards the misleading marketing, journalism, and presentations about D-Wave projects.

How many of those use the D-WAVE machine in production?

Zero.

Everyone are just trying to work out if they can use it for something - anything. So far the answer is NO.

Several universities and corporations have QC R&D projects (eg. Alibaba, Google, IBM, Microsoft); none of these corporations have customers for their QCs.

This is what LANL has to say...

http://www.lanl.gov/discover/publications/1663/2016-july/not...

...that went over my head depressingly quickly....

Many of the points made why successful organizations invest in new technologies are very weak. Successful organizations got that way by making many more smart decisions than dumb decisions.

Occam’s razor suggests that organizations investing in adiabatic quantum computing do so for reasons the participants of this blog don’t have access to.

One thing I'm never quite sure of in these discussions as to what extent D-Wave are a bunch of shysters and to what extent they're actually honest believers that they've built a quantum computer and that their repeated savagings are nothing more than the reasonable expectations of sceptical science being done.

Basically do they really believe they've built a quantum computer or do they actually know they haven't and have decided to pretend?

I have a feeling what's happened is that they built something they thought was quantum, but since it's so hard to measure qubits they had to develop a bunch of powerful statistical models to help read the qubits.

These models have gotten so good at reading "correct" results from qubits that they've actually started solving the problems themselves.

So are the qubits actually doing anything now or are the models solving problems from totally random qubit inputs? I think that's a question that even they can't answer

I wonder if this was another case of engineering/marketing disconnect, and by the time everything got sorted out they were in over their head.

Engineer: "It can use supercooled qubits to solve certain kinds of annealing problems and-"

Marketing (and management): "It's a quantum computer?"

Engineer: "Well, no, it sort of uses quantum mechanics to to operate, but it doesn't-"

Marketing: "For immediate release: D-Wave builds quantum computer!"

In this sense we can argue that Intel will perhaps soon deliver a quantum computer since Intel is currently researching on 10nm Quantum Well Field Transistors (QWFET) to replace FinFET transistors that are currently used:

> http://www.digitaltrends.com/computing/analyst-predicts-inte...

If we are evil marketers we can even claim any computer is a quantum computer (though not using quantum effects in an essential way - but this can be left out), since anything in the world (e.g. matter) only exists because of the laws governing quantum physics.

If you expand your criteria to unaffiliated evil clickbait authors, you'll surely find that, and worse. The bullcrap I've seen shared on transhumanist/AI Facebook groups is just mind-boggling.

Living in Vancouver, I know several people who have/do work at D-Wave, and I am certain that they are not a company of shysters. There are a lot of smart people working to make the thing do the thing, and I think the majority aren't concerned with proving it at every step, they're just trying to make it happen (because they believe that it's possible given current knowledge and technology).

It wouldn't surprise me if there's a bit of a PR / Tech disconnect.

The smartest human I have ever known (a guy I went to university with) worked for them for a few years. He's the kind of person that if he says they're legit, I believe it because he's too smart for them to have fooled him, and he's too honest to be part of a fraud. He never gave any indication that he thought it wasn't legit.

As far as I can tell, they think they're using quantum effects to speed up computation. The two questions that remain are A: are they right? and B: are the effects they are using giving them enough of an advantage to beat classical computers?

Carl Sagan has written at length about the limits of the credibility-assessing heuristic you're following. There's a reason that scientists have been snowed by paranormal fraudsters while professional magicians have not. I know nothing about D-Wave and have no opinion, but you need to look to people who manipulate people for a living—think the Amazing Randi—for accurate assessments of these sorts of things, not scientists. To paraphrase Sagan, the problem with scientists is that they are trained to fight against an opponent—Nature—that typically fights fair.

If you think scientists in academia fight fair, I can only look on your Persil white view of the world with envy.

Looks to me they are composed of:

1 - Some engineers honestly working on a cool new device that can maybe improve computation;

2 - Dishonest salespeople trying to hype their product on any way that would increase sales.

It started pretty much honest, with real explanations of their machines, and more research based deals than commercial ones. Then, when it started to market for government(s) their computer become a "quantum computer!!!" and everything they said stopped making sense.

I think it doesn't matter that much to most people... only the results matter. You can really hear that in the blog we're responding to here. There's no need to worry about motives and all of the other bullshit, they just need to make their device do what it's supposed to, in a way that can't be replicated by a similarly dedicated classical computer.

If they're people with the best of intentions and stars in their eyes, who produce nothing of value... who cares? If they're cynical hacks working their way up the funding ladder until they can actually produce what they've claimed to have for years, and actually produce it? Great, again, the backstory matters a lot less to most who aren't looking to invest.

D-Wave existing doesn't prevent anyone else from pursuing quantum supremacy, and no one is being forced to invest in something so dubious.

Perhaps better to link to the blog post than the blog index: http://www.scottaaronson.com/blog/?p=3192

Thanks! We've updated the link.

tl;dr Still no evidence that D-Wave's new "thousands of times faster" computer is any faster than a classical computer, even for annealing tasks. And we'll see if we've reached quantum supremacy when/if Google and IBM reveal their 50-qubit quantum computers in the next few years.

Sad to see a blogger I otherwise like spouting the "Trump is a Russian puppet" propaganda.

This is not the first attack against Trump from Mr. Aaronson. http://www.scottaaronson.com/blog/?cat=42 Also it seems he has a more than cozy relationship with YC so watch your negative comments about him on this site!!

I'm not sure if it's virtue signaling or what--this tendency to inject "by the way, I'm anti-Trump" into articles that have nothing to do with that subject. It's annoying though.

Scott Aaronson is a smart guy, but I care about his politics about as much as I care about Matt Damon's.

This sure sounds like sour grapes to me, arguing that a faster, or true quantum computer will be built in a few years that will out preform Dwave's current system is pretty weak sauce. After reading this article I can't help but see the similarities between this article and the recent attack posts against uber that seem to be the flavor of the day. DWave has built and sold several functioning systems and this is what most likely puts a bullseye on them for people to take cheap shots..just like the Uber haters. Regardless of whether or not DWave has built a true quantum computer, or if the next conventional computer will be faster, they are doing it now, they have found the more effective algorithms and they are pushing the envelope. It's always easier to criticize and whine than stick your neck out and try something new and innovative and articles like this one show that even if you do succeed in those endeavors the haters are going to hate. The one part of the article that I did appreciate is the mention of Geordie Rose who was one of DWave's founders and the fact that an ex wrestler founded such a cutting edge company may speak volumes to the need for strong willed people in unicorn type companies to fend off then nay sayers before they even get off the ground.

I wonder why this is heavily downvoted. I am not a scientist but I have followed D-Wave news since a few years and they seems to be a serious company. They are not making huge claims either, no free energy, Mars colonisation in two years, or global Internet with free balloons. It seems to me that there is currently a need for some new computing architectures as silicon encountered its speed limit in 2011 or so. There are even blogs that are dedicated to this subject [0]. To see innovation emerge, we need some randomness, fools trying new things, failing often and be forgotten. Without them I would not be able to read HN online and type on a computer keyboard.

[0] https://www.nextplatform.com/

Nobody, not a single company, is using D-Wave in production for anything. It's not a useful machine, it doesn't do anything you can't do with a much cheaper regular computer.

Aaron knows what he's talking about, the criticism is not 'sour grapes'.

Several universities and corporations have QC R&D projects (eg. Alibaba, Google, IBM, Microsoft); none have their QC machines in production for anything (in fact they don't even have customers). Something to think about.

Reply on news.ycombinator.com