Slashdot Mirror
First Demonstration of Artificial Intelligence On a Quantum Computer
KentuckyFC writes: Machine learning algorithms use a training dataset to learn how to recognize features in images and use this 'knowledge' to spot the same features in new images. The computational complexity of this task is such that the time required to solve it increases in polynomial time with the number of images in the training set and the complexity of the "learned" feature. So it's no surprise that quantum computers ought to be able to rapidly speed up this process. Indeed, a group of theoretical physicists last year designed a quantum algorithm that solves this problem in logarithmic time rather than polynomial, a significant improvement.
Now, a Chinese team has successfully implemented this artificial intelligence algorithm on a working quantum computer, for the first time. The information processor is a standard nuclear magnetic resonance quantum computer capable of handling 4 qubits. The team trained it to recognize the difference between the characters '6' and '9' and then asked it to classify a set of handwritten 6s and 9s accordingly, which it did successfully. The team says this is the first time that this kind of artificial intelligence has ever been demonstrated on a quantum computer and opens the way to the more rapid processing of other big data sets — provided, of course, that physicists can build more powerful quantum computers.
Now, a Chinese team has successfully implemented this artificial intelligence algorithm on a working quantum computer, for the first time. The information processor is a standard nuclear magnetic resonance quantum computer capable of handling 4 qubits. The team trained it to recognize the difference between the characters '6' and '9' and then asked it to classify a set of handwritten 6s and 9s accordingly, which it did successfully. The team says this is the first time that this kind of artificial intelligence has ever been demonstrated on a quantum computer and opens the way to the more rapid processing of other big data sets — provided, of course, that physicists can build more powerful quantum computers.
Crap. now what are we going to do instead of using a captcha?!
And also no working quantum computer, except for very small toys. Pattern recognition is not AI.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
I read this:
And immediately thought of this:
God, I love how weird the future is.
Koans and fables for the software engineer
Would. You. Like. To. Play. A. Game?
Brace yourself, quantum algorithm classes are coming.
What they actually did if you read the paper is:
1) Encode a 6 or 9 image into 2 numbers, based on the number of excess pixels in the left vs right, and top vs bottom quadrants. From the article: After these preprocessing, the two printed image with standard font can be represented by ~x1= (0:9872;0:1595) for character "6" and ~x2= (0:3544;0:9351) for character "9"
2) Use a training algorithm to find the appropriate pulse sequence to give a up result from the molecule's NMR C13 spectra from a 6, and a down signal from a 9.
3) Run the NMR spectrum, feed in pulses based on the parameters produced from pixels encoded in a vector form like 1), get the result of "up" for a 6 and "down" for a 9.
It's certainly neat experimental NMR work, but I don't really see how it's quantum computing. But then maybe that's the NMR spectroscopist in me talking....
(I can't believe I didn't notice I misspelt Quantum in that subject field.)
When will you become self aware?
Feed it crapchas until it catches on!
When you live in a sick society, just about everything you do is wrong.
It's smart and not smart at the same time.
Table-ized A.I.
Can we get a link to a more reputable site?
>There's no 'intelligence', there's fancy pattern recognition.
IQ tests are heavily based on (timed) pattern recognition.
>Vending machines have been able to identify what kind of coin you put in for decades. That doesn't make them 'intelligent'.
Why not?
>I have no idea of the formal definition of AI,
That would be a problem.
> but to me without some form of abstract decision making and actually applying it to something, it's just clever automation
It's taking a real-world pattern and deciding how to abstract it.
Oh... I was interested in this, and then saw it was Medium.com
Meaning it would be overhyped nonsense and have nothing to do with the title of the story.
Saddly, I was interested enough to read through it and prove my "Medium.com sux" theory is still correct.
But you forgot to include the word "imitate" in your highlight. Pattern recognition is most definitely imitating an intelligent human behavior.
Well, I for one welcome our new Quantum overlords.
Ho.
Ly.
Shit.
15 or 20 years ago, I was saying that because quantum computers perform multiple calculations on similar inputs simultaneously, they'll be perfect for the sorts of pattern recognition tasks needed for (artificial) intelligence. And now these smart people have figured out how to do it for the first time, albeit with a miniscule 4 qubit quantum computer.
But since quantum computing capabilities scale according to 2^n, where n is the number of qubits, a 24 qubit computer (i.e. 6 times the size of what they just built, requiring a molecule with 24 atoms) would be 2^20 = 1 million times as powerful as this 4 qubit computer just demonstrated. A 64 qubit computer would be 10^18 = 1 million million million times as powerful as this 4 qubit computer. Good-bye conventional computer encryption. And hello general-purpose pattern-recognition (i.e. the basis for strong artificial intelligence).
My first thought was that a vat of "carbon-13-iodotrifluroethylene" isn't exactly a general purpose computing device -- except that because their control inputs are a stream of radio waves pulses controlled by a conventional computer, it actually is a general purpose computer. And though I'm no quantum physicist / quantum computer scientist, it seems like it would scale reasonably easily: you just need to find larger organic molecules with similarly discrete nuclear magnetic resonance 'channels' (i.e. independently manipulable/separable by frequency).
I am beginning to sense the coming Kurzweil Singularity...
That's OK, you also misspelled "exaggerating", so I didn't notice ;-)
Mostly, yeah, it's pretty exaggerated as AI. It's potentially an interesting piece of work, but I always get skeptical when the PR departments feel they have to exaggerate.
Does the this mean that AI is only 20 years away?
The /. summary says "The computational complexity of this task is such that the time required to solve it increases in polynomial time with the number of images in the training set and the complexity of the "learned" feature." Moore's Law is such that any polynomial time problem will be trivially solved by the exponential advances of Moore's Law. If this problem were exponential in nature, not polynomial, then quantum computing might be our only hope. But polynomial-time problems are not the sweet spot for quantum computers.
(I can't believe I didn't notice I misspelt Quantum in that subject field.)
If you want to, we have this neat quantum computer which is capable of making the difference between 'n' and 'h'...It might be pricey but it is totally worth it...
It occurred to me today that the fundamental question regarding the morality of AI in society is as follows (aside from the fact that it may be inevitable):
Is it possible to dehumanise/machine a human to be less or equally "moral" than a machine.
Is it possible to make a machine that is more or equally "moral" to a human?
That's really a never ending argument but for me, but it seems that if you had both of these things above, you would have a machine that is more morally correct than a human. You could put a lot of care into a machine and make it more morally correct than an ISIS rebel beheader.
The problem is not out technology, as portrayed in scare mongering movies like Terminator.
The problem, as always, is us. Machines could only be bad if we make them bad.
You could hardcode unbribeable integrity into a machine. You can't do that with a human. Humans can be broken, even if they snap just out of a pure 3rd. person out of body compassion for themselves as a human being, we can snap, or be corrupted if the chips are down, enough pressure.
I was very skeptical of AI when I was young. I was brainwashed by Hollywood - that's all it is brainwashing. Technology can free people from mundane tasks and bring a level of fairness to society we could only dream of. Always fair and unbiased.
I don't know if most humans will benefit from this, as there could be some mass outbreak like Ebola, economic/bankster catastrophes, food shortages, technological mistakes etc. that will curtail our present civilisation, but in the far distant future, I think the best way is for machines to run our lives. Machines, that are not human, but still very advanced intellectually. Not human at all. Humans aren't evolved enough to lead themselves into the future. We're evolved to survive and protect or families, our tribes. For the survival of a civilisation long term, you need something that's non human I feel.
They took and existing technology and ran an existing algorithm on it to solve a trivial problem to prove an idea that was never in doubt.
Clearly this demonstration was designed by someone who was searching for the deeper meaning of the work of Jimi Hendrix:
http://www.youtube.com/watch?v=eyGWbpNzH2o
When are we gonna hollow out Deimos and fly to Tau Ceti?