Friday, December 18, 2020

Latest Quantum Supremacy claim starts to crumble

Scott Aaronson backtracks on quantum supremacy. After telling us for years that Gil Kalai was wrong to be skeptical about quantum supremacy, he now admits that one of Kalai's criticisms turned out to be correct.

There are now two high-prestige publications claiming quantum supremacy. Aaronson was the referee on both, and thus he got to decide that these papers were worthy of the claim.

Here is the game. Some quantum researchers do a complicated experiment, and then declare that it would be hard to simulate. But no one is particularly interested in simulating it, and we don't know how hard the problem is.

An example of a hard quantum problem is protein folding. Proteins naturally fold in milliseconds, but researchers have spent decades trying to find algorithms to predict foldings. If the protein were a quantum computer, then Aaronson would say that it has acheived quantum supremacy.

Except that Google DeepMind may have just cracked the problem. This work is very exciting, and may be worthy of next years Nobel Prize in Chemistry. They did not use quantum computers.

So it does not show quantum supremacy at all.

Maybe all these quantum supremacy experiments could be simulated by classical computers, if a team of experts spent 30 years finding a good algorithm.

Ultimately someone may find a quantum experiment that is hard to simulate. But in my mind, it won't really be quantum computation unless it computes something like Shor's algorithm that cannot be in reasonable time on a classical computer.

Update: Also amusing is Aaronson's new blog tagline:

If you take nothing else from this blog: quantum computers won't
solve hard problems instantly by just trying all solutions in parallel.

"The Far Right is destroying the world, and the Far Left is blaming me!"
The first part has been there for a while, and it refers to the facts that nearly everyone describes quantum computers as trying solutions in parallel, but they have never been shown to do that.

The new part refers to him agreeing with 95% of the Leftist agenda, but that makes him seem like a right-winger in today's academia.

No comments:

Post a Comment