Google has big plans for quantum computing.
The company has come up with a strategy for demonstrating quantum supremacy, the claim that quantum computers can perform tasks that no current computers can. While it’s widely assumed that we will eventually reach quantum supremacy, nobody has done it yet because current quantum computers can only run a small number of specialised algorithms.
Their plan is based on simulating coin flips. An ordinary computer does this by storing two numbers and choosing one of them at random each time. To simulate 50 coin tosses, it just selects 50 times in a row.
This is simple with regular coins, but if the coins behave like particles obeying the laws of quantum mechanics, things get more complicated.
In that case, we cannot know whether any individual coin turned up heads or tails without knowing about all the other coins, a phenomenon known as quantum entanglement. The problem of simulating coin tosses with quantum entanglement is called quantum sampling.
Computers work sequentially, so choosing 50 numbers at the same time is not something they can do. For this reason, the Google group argues, quantum sampling would require storing all possible configurations of all 50 coin tosses, so that all of the coins can be thrown simultaneously.
Since one bit – the building block of classical computers – can only store one of two states, heads or tails, covering all possible configurations for 50 coins takes hundreds of terabytes of data storage.
This is where quantum computers come in. They’re based on qubits, which can be in two states at the same time. This makes it possible to store the probability distribution of all the configurations at once using a single qubit for each coin. For this reason, the Google group argues, quantum sampling would be easy for a quantum computer.
In their proposal, the team demonstrates quantum sampling up to nine coins with high accuracy using their 9-qubit quantum computer. “If similar error rates are achievable in future devices with around 50 qubits, we will be able to explore quantum dynamics that are inaccessible otherwise,” the proposal states. This way, quantum computers of the near future can be used to study physics, a huge step-up from their infancy when they couldn’t do anything practical.
The only remaining task, then, is to build a 50-qubit computer. And the team just might, given their track record
“They’ve been delivering many of the expectations,” says Scott Aaronson at the University of Texas in Austin. “The truth is, the Google group has such a strong record that if they say they’re going to do it, people pay attention.”
Indeed, the Google group has completed the necessary preparatory work to build large-scale quantum computers, according to Jacob Taylor of University of Maryland. “They’ve demonstrated that the obvious pitfalls are largely accounted for,” he says.
Pick your problem
Not everyone is convinced quantum sampling is the right problem to tackle in order to demonstrate quantum supremacy, though.
“It is unclear whether what they claim to show is quantum supremacy,” says Itay Hen of the University of Southern California. “You have to prove that classical computers can’t do it.”
He says that many important quantum-mechanical systems can be simulated on present-day computers, because it’s not necessary to save all of the information about the system to simulate it.
Proof that this kind of simulation can’t be done on a classical computer may not come anytime soon, but many concepts in this field are accepted as true without formal proof.
“We know almost certainly there’s not going to be a fast classical algorithm” that solves quantum sampling problems, says Aaronson. If there isn’t, and if Google manages to put their plan in motion, it may finally be a hint that quantum computers really are better than regular ones.
Source: New Scientist