[ad_1]
Scientific advances do not always go as planned. On Friday, a draft research paper mistakenly downloaded from NASA's website accidentally revealed to the world that Google had reached a milestone in quantum computing: quantum supremacy.
This is a goal that Google – and its competitors – have been aiming for for years. In 2017, the firm predicted that it would reach quantum supremacy by the end of the year in question, but this deadline has not been reached. In the years that followed, IBM and Intel piqued Google's heels, testing quantum computers with an ever-increasing number of qubits – the information units that explain the potential power of quantum computers.
It seems now that Google has taken this step ahead of its competitors. The document describes in detail how Google's researchers used a quantum processor called Sycamore that contained 53 qubits in working order to solve a random sampling problem that would have taken 10,000 years to the world's best supercomputers. Sycamore took only three minutes and 20 seconds. Google, which has partnered with NASA for this project, has not responded to requests for comment.
But this breakthrough does not mean that useful quantum computers are at hand. Not far away. Instead, Google has just opened the door to the next era of quantum computing. And that's where things start to get really interesting.
Quantum supremacy is not as exciting as it sounds, says Simon Benjamin, professor of quantum technologies at Oxford University. It simply means the moment when a quantum computer completes a task that conventional computers find impossible. In the case of Google, it meant telling the quantum computer to run a set of random instructions and then measuring the results. The researchers then tried to make sure that a supercomputer could predict what the quantum computer would produce, to make sure that the results were only realizable by a quantum computer.
This task is virtually useless. It's good to sort out the quantum computers of their classic parents, but that's it. And that means that, in some ways, quantum supremacy is also useless. "His [a milestone] it must be spent to get very good results, but it will not have any immediate consequences, "says Benjamin.
Some researchers, including Benjamin, believe that the term quantum supremacy, popularized by the theoretical physicist John Preskill, is a bit exaggerated. Benjamin has a suggested alternative – quantum inimitability – which, he says, gives a little more precision to the technical aspects of supremacy, without inviting any comparison between Terminator and its designers.
But the simple fact that the actual task used by Google to demonstrate supremacy is meaningless does not mean it does not matter. "[The supremacy test] is intended to give you greater confidence that this machine is performing computer tasks that can not be reproduced on a typical computer, "says Toby Cubitt of the Quantum Science and Technology Institute at University College London.
In the race for issues to make quantum computers work
This is great news and a huge seal of approval for quantum computing. In October 2017, IBM demonstrated that it could simulate quantum computers with 56 qubits on non-quantum supercomputers. And if you can simulate quantum computers, why not bother to create a device the size of a room that needs to be cooled to the limit of an absolute zero?
Google's paper displays a massive flag reminding everyone that there is a whole class of calculations that can only be solved by using quantum computers, and opens the door to those who would really like to start designing problems at home. solve by these computers.
But what comes next is much more exciting. Google's experience draws a line in the sand between two epochs of quantum computing. Welcome to the era of supremacy. In front of us, we call something called intermediate-scale quantum noise (NISQ). If the ultimate goal of quantum computing is to be able to perform calculations that no conventional computer could ever dream of performing – complex molecular interactions or cryptographic code deciphering – then NISQ is like an awkward quantum childhood.
The problem is that even the most promising quantum computers are disastrously prone to errors. Imagine asking a computer to perform a calculation and that once a week or more, you just have to give a totally random answer. And this, calculation by calculation: a program can consist of several millions of individual calculations. Google's Sycamore processor would not be able to tackle any useful quantum problem.
The NISQ era is about making quantum computers that work pretty well, despite all that noise. Benjamin explains that the solution is to find a way to make the most of imperfect machines. This could mean engineering programs simple enough to run on quantum computers while producing useful results. And the other part will consist of improving hardware so that quantum computers produce useful results while using only a relatively small number of qubits.
Benjamin has a name right now, we take on the challenge of making noisy quantum computers that are really useful: the quantum advantage. "[It’s about] how we can take advantage of imperfect machines, "he says. "At least we will be at a time when quantum computers have started to become useful tools."
And if we have gone beyond the quantum advantage, then we can start thinking about the big challenges that only quantum computers can solve, such as the factorization of very large numbers or the modeling of quantum mechanics. This will require another type of quantum computer, a computer that will solve all its errors and use millions of qubits to perform calculations.
Although Google may have taken the lead, there is no way to know if its quantum computer – based on superconducting electronic circuits – will form the basis for future useful quantum computers. "It's hard to predict what hardware will lead to a million-bit computer," said Natalia Ares from Oxford University. In the early years of quantum computing, ion traps were the most common type of hardware, but they were later overtaken by the superconducting machines preferred by Google and IBM.
For Ares, the news from Google is exciting, but it is important not to wait for ordinary quantum computers useful. The biggest quantum computer to date is Google's Bristlecone 72-bit processor, which has not even demonstrated quantum supremacy. We are very far from the millions of qubits we will need to solve difficult quantum problems. "We are far from doing anything very useful but it shows that we are making progress," said Ares. Google may have made some progress, but the marathon is just beginning.
More beautiful stories from WIRED
? Tesla created a battery that could last a million miles
?️ Four-day work weeks are not a utopia, it seems
? Japanese self-cleaning toilets conquer the West
The foods you really need to stock for a Brexit without agreement
Get the best tech deals and the latest gadget news in your inbox
[ad_2]
Source link