Quantum computers won’t be truly useful until they can correct their mistakes
davide bonaldo / Alamy
Quantum computers are already here, but they make far too many errors. This is arguably the biggest obstacle to the technology really becoming useful, but recent breakthroughs suggest a solution may be on the horizon.
Errors creep into traditional computers too, but there are well-established techniques for correcting them. They rely on redundancy, where extra bits are used to detect when 0s incorrectly swap to 1s or vice versa. In the quantum world, however, it is a lot more challenging.
The laws of quantum mechanics forbid information from being duplicated inside a quantum computer, so redundancy must be achieved by spreading information across groups of qubits – the building blocks of quantum computers – and utilising phenomena that only exist in quantum settings, such as when pairs of particles become linked via quantum entanglement. These qubit groups are called logical qubits and figuring out the optimal way to build and use them is crucial for determining how best to eliminate errors.
A recent surge in progress has made researchers optimistic. “It’s a very exciting time in error correction. For the first time, theory and practice are really making contact,” says Robert Schoelkopf at Yale University.
One of the stumbling blocks for quantum error correction has been that the number of qubits needed to make a logical qubit tends to be large, which makes the whole quantum computer costly and challenging to build. But Xiayu Linpeng at the International Quantum Academy in China and his team have recently demonstrated that this doesn’t have to be the case.
The researchers found that just two superconducting qubits can be combined with a tiny resonator to make one larger qubit that both makes fewer errors and can automatically flag an error when it happens. They then went a step further to show how three such qubits can be grouped together through quantum entanglement for building up computational power without surreptitious errors.
Schoelkopf’s team also recently demonstrated how several operations necessary for quantum computer programs could be implemented with the same type of qubit and exceptionally low error rates, with some errors occurring as rarely as once in a million qubit manipulations.
Even though approaches like this will catch many errors, useful quantum computers will have to contain thousands of logical qubits, meaning some will still creep in. So Arian Vezvaee at start-up Quantum Elements and his colleagues have tested a way to add further error protection to logical qubits, like wearing a raincoat under an umbrella.
The key idea is to not let any qubits sit idle for too long, as that makes them lose their special quantum properties and become corrupted. The team showed that giving idle qubits extra “kicks” of electromagnetic radiation can create the most reliable entanglement between logical qubits to date.
The exact recipe for how to combine physical qubits into logical ones really matters for some of the most precise calculations, as David Muñoz Ramo at quantum computing firm Quantinuum and his colleagues found when investigating an algorithm that determines the lowest possible energy that a hydrogen molecule can have. There, the precision needed is so high that basic error-correcting methods aren’t enough.
Such innovation in error-correcting programs will be crucial for the success or failure of quantum computers, says James Wootton at start-up Moth Quantum. “We’re still in a phase where researchers are learning how all the pieces of error correction fit together.” Quantum computers can’t yet operate effectively without errors, but we are starting to see the engineering foundations of this appear, he says.
Topics: