Recent news has thrust the already burgeoning field of quantum computing to the forefront of trending technology topics. In mid-November, Google announced via a paper in the journal Nature that along with its partners at the University of California-Santa Barbara it had achieved a milestone in computing, performing the first demonstration of quantum supremacy.
While the media was quick to pick up the tale being told by Google, others—most notably IBM, which is also competing to dominate the quantum computing space—disagreed with Google’s proclamation, dismissing the announcement as a publicity stunt.
So who’s right?
Did Google achieve quantum supremacy? Is IBM right or is its rebuttal just for the sake of argument? Does it matter?
To answer that, maybe it’s best if we start at the beginning.
What Is Quantum Supremacy?
Quantum supremacy, as first defined by John Preskill in his 2012 paper Quantum Computing and the Entanglement Frontier, is the state in which a quantum computer can outperform a classical computer by a superpolynomial time factor.
In the current case of Google and IBM, Google claims to have used a 53-qubit quantum computer to solve a problem in mere minutes that would have taken a classical computer millennia to calculate.
“A computation that would take 10,000 years on a classical supercomputer took 200 seconds on our quantum computer,” said Brooks Foxen, a graduate student researcher at UC-Santa Barbara.
However, Foxen did add a caveat to his description of the Google announcement.
“It is likely that the classical simulation time, currently estimated at 10,000 years, will be reduced by improved classical hardware and algorithms, but, since we are currently 1.5 billion times faster, we feel comfortable laying claim to this achievement,” he said.
IBM Rebuts
Before Google or Nature ever got to press, IBM was already beginning to challenge Google’s claim. On Oct. 21, two days before Nature would run the Google/UC-Santa Barbara paper, IBM posted a blog asserting the most boastful claims being made by the researchers were over-hyped.
Researchers at IBM argued that “an ideal simulation of the same task [that Google used as its problem set for this announcement] can be performed on a classical system in 2.5 days and with far greater fidelity.” IBM went even further, stating that its two and a half day estimate was a rough estimate and that with further tinkering, “the classical cost of the simulation can be further reduced.”
IBM said that the Google paper cherry-picked their results. In a particularly dense portion of its blog, IBM engineers said that Google engineers simplified how a classical computer works by limiting the resources it could use to solve the problem they set their quantum computer upon:
“[Google’s] classical simulation estimate of 10,000 years is based on the observation that the RAM memory requirement to store the full state vector in a Schrödinger-type simulation would be prohibitive, and thus one needs to resort to a Schrödinger-Feynman simulation that trades off space for time.”
IBM then broke down the high-level computational physics explanation to something understandable by the public:
“The concept of ‘quantum supremacy’ showcases the resources unique to quantum computers, such as direct access to entanglement and superposition. However, classical computers have resources of their own, such as a hierarchy of memories and high-precision computations in hardware, various software assets, and a vast knowledge base of algorithms, and it is important to leverage all such capabilities when comparing quantum to classical.”
Basically, IBM is saying that while Google’s machine performed like Usain Bolt, its competition was a one-legged man.
The Problem with Quantum Supremacy
When IBM wasn’t dissecting Google’s PR move, it was complimentary of Google’s commitment to developing quantum computers and advancing a new type of calculating machine.
“Building quantum systems is a feat of science and engineering and benchmarking them is a formidable challenge. 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.”
IBM added, “But it should not be viewed as proof that quantum computers are ‘supreme’ over classical computers.”
Though that turn of phrase could be misconstrued, it’s more of a course-correction intended for the entire conversation occurring around quantum computing since the Google announcement.
Quantum computers are not classical computers. They solve problems in completely different ways using different physics. Classical computers like today’s most powerful supercomputers are physically robust and very good at solving massive problems, such as weather simulations and the physics inside of a nuclear reactor—provided the problems being solved can be broken down into smaller bits, solved and then unified into a coherent whole. The difficulties with classical computers are that they are physically massive, energy intensive and are verging on the limit of how powerful they can be dueto chip manufacturing constraints bounded by nanoscale etching and electrical properties.
Quantum computers are in their infancy, and they’re already proving to be very powerful. However, they are very delicate objects. Because of their quantum nature, quantum computers have to be completely isolated from the surrounding environment to achieve accurate results—analogous to Schrödinger's cat. This makes it difficult to realistically imagine quantum computers becoming dominant over classical computers for widespread use. With our current and expected near-future technology, they’re too hard to maintain.
In all likelihood, quantum computers will serve as a complement to classical computing architectures, jumpstarting and advancing a problem to a point where a powerful classical computer can take over and do some serious number-crunching.
Where Does This Leave the Supremacy Debate?
Currently, the supremacy debate seems to be largely the domain of marketers. ‘Supremacy’ is a somewhat meaningless term in many respects, especially when discussing specialized technology.
One day a quantum computer will likely out-computer a fully-functional classical computer when solving a problem, but that won’t be the demise of classical computers. A calculator is still useful when doing bills even with a workstation right beside you; but you would turn to the workstation when it’s time to run a model.
Assessing supremacy boils down to an analysis of need and use. Because of that, both quantum computers and classical computers will find themselves the more valuable at the right moment.