Google researchers today published their latest work on quantum computing in Nature, showing off how its new Sycamore processor can run a test computation in 200 seconds that would take the world's biggest supercomputers 10,000 years to complete.
News of the paper leaked via a Nasa website last month, detailing that Google had achieved so-called 'quantum supremacy', which happens when a quantum computer can solve problems that would take a classical computer too long to be considered practical.
IBM this week challenged Google's claim that it had achieved quantum supremacy, because the ad giant's researchers failed to account for "plentiful disk storage" and other assets on a classical computer. Rather than 10,000 years, IBM researchers claim Google's challenge would take a classical computer just two and half days.
"Because the original meaning of the term "quantum supremacy," as proposed by John Preskill in 2012, was to describe the point where quantum computers can do things that classical computers can't, this threshold has not been met," IBM researchers wrote.
IBM recently announced its 53-qubit system should be available in mid-October, so it has a significant interest in not being leap-frogged by Google's work.
"Google's experiment is an excellent demonstration of the progress in superconducting-based quantum computing, showing state-of-the-art gate fidelities on a 53-qubit device, but it should not be viewed as proof that quantum computers are 'supreme' over classical computers," the IBMers said.
Nonetheless, Google CEO Sundar Pichai has heralded Google's work as a "big breakthrough in quantum computing known as quantum supremacy".
For the geeks out there, Pichai described it as "the 'hello world' moment we've been waiting for – the most meaningful milestone to date in the quest to make quantum computing a reality".
The allure of quantum computing is that a qubit can be both a 0 and 1 simultaneously thanks to the quantum property called superposition.
So, instead of just a 1 and 0 in classical computers, the 1 and 0 can on a quantum computer be in four possible states at any time. A quantum computer with 54 qubits can have 2^54 computational states and since it can scale exponentially, it holds the possibility for computers to one day solve much more complex challenges.
Google researchers explain the Sycamore processor in more detail on the company's AI blog. A key achievement is that the group addressed quantum computers' tendency to be error prone.
"It's comprised of a two-dimensional grid where each qubit is connected to four other qubits. As a consequence, the chip has enough connectivity that the qubit states quickly interact throughout the entire processor, making the overall state impossible to emulate efficiently with a classical computer," the Google researchers explain.
The quantum supremacy experiment was enabled by its improved two-bit gates, which are the foundation of quantum circuits and serve the same function as logic gates in classical computers. Google showed it could achieve record performance even when operating many gates simultaneously.
"We achieved this performance using a new type of control knob that is able to turn off interactions between neighboring qubits. This greatly reduces the errors in such a multi-connected qubit system. We made further performance gains by optimizing the chip design to lower crosstalk, and by developing new control calibrations that avoid qubit defects."
As noted in a Nature report on the research, Google's supremacy demonstration was carried out by "sampling solutions" on a circuit implemented on Sycamore. The results were then compared with simulations performed on classical supercomputers, including the huge Summit supercomputer at Oak Ridge National Laboratory in Tennessee.
Google says in the future it will make its "supremacy-class" processors available to collaborators, academic researchers and companies that want to develop algorithms and applications for today's quantum processors.
The group is also racing to build a fault-tolerant quantum computer that they believe can be used to design new materials, such as lightweight batteries for cars and airplanes, better fertilizers and more effective medicines.
"Achieving the necessary computational capabilities will still require years of hard engineering and scientific work. But we see a path clearly now, and we're eager to move ahead," the researchers noted.