October 24, 2019 by Luke James
Google attempts to establish "quantum supremacy" by unveiling its landmark 'Sycamore' chip, a quantum computing chip now being prepared for preliminary electrical testing.
A Google quantum computer named 'Sycamore' has achieved quantum supremacy by outpacing ordinary computing technology. On Wednesday, October 23, Google disclosed its data processing milestone in the Nature journal. Google's "quantum supremacy" achievement marks more than a decade's worth of work at Google.
In a blog post regarding their work, Google researchers said that "Our machine performed the target computation in 200 seconds, and from measurements in our experiment, we determined that it would take the world's fastest supercomputer 10,000 years to produce a similar output,".
By completing its target computation, Google's Sycamore supercomputer achieved "quantum supremacy" and solved a problem that has been too hard for the world's best supercomputers to crack and would take them 10,000 years according to estimates.
Left: An artist's rendition of the Sycamore processor mounted in the cryostat. Right: A photograph of the Sycamore processor. Image courtesy of Google from Google AI Blog.
Despite criticism from some that claim that Sycamore has done nothing but solve a single, contrived problem, and that Sycamore does not yet do anything worthy of note, Google scientists are hopeful and believe that they have achieved a significant computing milestone.
Google engineer Hartmut Neven said, “One criticism we’ve heard a lot is that we cooked up this contrived benchmark problem—[Sycamore] doesn’t do anything useful yet…. That’s why we like to compare it to a Sputnik moment. Sputnik didn’t do much either. All it did was circle Earth. Yet it was the start of the Space Age.”
As you may well know, today's computers use transistors to represent data as binary code with zeroes and ones. Quantum computers, on the other hand, represent data using quantum bits (qubits), an artificial atom that is capable of holding a state of superposition.
Rather than use rules of logic, qubits interact using quantum mechanics. They take on zero or one and produce binary code just like current computers do, however, each qubit can store a combination of different states of zero and one at the same time.
The Sycamore processor and its 54 qubits, one of which is not working correctly. Image Credit: Nature Journal.
Multiple qubits can be connected via entanglement and this lets a quantum computer concurrently explore a huge number of potential solutions to a given problem. This means that the more qubits you have, the more powerful your quantum computer is and engineers are working hard to increase the number of qubits they can pack into their machines (just like transistors on a chip!)
Google's Sycamore processor had 54 qubits, however, only 53 were available because one was not working correctly.
There is still a long way for Google to go before its quantum computer can be considered viable. At present, lots of research is required before the true power of quantum computing can be realized.
Additionally, there is still lots of error correction to be done by Google researchers, something that will likely require new technological breakthroughs. By improving qubits to produce fewer errors, there is the potential for more qubits to be interlinked which will bolster computing power but, at the same time, create other problems such as the potential for overheating.
While exciting and very different from traditional computing, quantum computing is not going to be transforming the tech space any time soon.
