The race to demonstrate “quantum supremacy” — a computational task possible for a quantum computer but not a classical computer — has dominated the media coverage of this exciting new technology. But while scientists debate the goalposts of this milestone (as well as its controversial name), some research groups explore a less flashy but still meaningful advance, “quantum advantage.” Here, the challenge is not to prove that a quantum computer can do something that no existing classical computer can pull off, but that the small-scale, noisy quantum computers of today can out-perform a similarly small-scale classical computer on a given challenge.
In Nature Physics this week, a group led by Sergey Bravyi of IBM Research published new findings that a “shallow” quantum algorithm can beat a similarly simple classical approach on a benchmark called a relation problem. This quantum speedup is limited — a more powerful classical computer can still solve it faster than the small-scale quantum device — but important for demonstrating that even today’s flawed quantum computers can have a leg up on their traditional peers, wrote Bill Fefferman in an invited perspective piece published by the journal alongside the research paper.
“I would call it an apples-to-apples comparison as compared with the apples-to-oranges comparison [of quantum supremacy],” said Fefferman, an assistant professor at UChicago CS. “They’re showing that for this new type of problem, they can get a really big speed up even with the noise.”
Read more at UChicago Department of Computer Science.
Image: “IBM 16 Qubit Processor” by IBM Research, licensed under CC BY-ND 2.0