A quantum device that detects and corrects its own errors

UC Santa Barbara researchers form partnership with Google
March 5, 2015

A photograph of the nine qubit device. Qubits interact with their nearest neighbors to detect and correct errors. (credit: Julian Kelly)

In what they are calling a major milestone, researchers in the John Martinis Lab at UC Santa Barbara have developed quantum circuitry that self-checks for errors and suppresses them — preserving the qubits’ state(s) and imbuing the system with reliability that is foundational for building powerful large-scale superconducting quantum computers.

“One of the biggest challenges in quantum computing is that qubits are inherently faulty,” said Julian Kelly, graduate student researcher and co-lead author of a research paper that was published in the journal Nature. “So if you store some information in them, they’ll forget it.”

Unlike classical computing, in which the computer bits exist on one of two binary (“yes/no”, or “true/false”) positions, qubits can exist at any and all positions simultaneously, in various dimensions. It is this property, called “superpositioning,” that gives quantum computers their phenomenal computational power, but it is also this characteristic which makes qubits prone to “flipping,” especially when in unstable environments, and thus difficult to work with.

Quantum error detection and correction scheme

Optical micrograph of the superconducting quantum device, consisting of nine Xmon transmon qubits with individual control and measurement, with a nearest-neighbor coupling scheme (credit: J. Kelly et al./Nature)

So the researchers developed an error process that involves creating a scheme in which several qubits work together to preserve the information, said Kelly. To do this, information is stored across several qubits.

“And the idea is that we build this system of nine qubits, which can then look for errors,” he said. Qubits in the grid are responsible for safeguarding the information contained in their neighbors, he explained, in a repetitive error detection and correction system that can protect the appropriate information and store it longer than any individual qubit can.

Key to this quantum error detection and correction system is a scheme called the surface code. It uses parity information — the measurement of change from the original data (if any) — as opposed to the duplication of the original information (as in error detection used in classical computing). That way, the actual original information that is being preserved in the qubits remains unobserved.

Avoiding decoherence

“You can’t measure a quantum state, and expect it to still be quantum,” explained postdoctoral researcher Rami Barends. The very act of measurement locks the qubit into a single state and it then loses its superpositioning power, he said. It’s akin to a Sudoku puzzle: the parity values of data qubits in a qubit array are taken by adjacent measurement qubits, which essentially assess the information in the data qubits by measuring around them.

“So you pull out just enough information to detect errors, but not enough to peek under the hood and destroy the quantum-ness,” said Kelly.

This quantum error correction has been proved to protect against the “bit-flip” error, but the researchers also plan on correcting the complementary error called a “phase-flip,” and also running the error correction cycles for longer periods to see what behaviors might emerge.

Since this research was completed, Martinis and the senior members of his research group have entered into a partnership with Google.


Abstract of State preservation by repetitive error detection in a superconducting quantum circuit

Quantum computing becomes viable when a quantum state can be protected from environment-induced error. If quantum bits (qubits) are sufficiently reliable, errors are sparse and quantum error correction (QEC) is capable of identifying and correcting them. Adding more qubits improves the preservation of states by guaranteeing that increasingly larger clusters of errors will not cause logical failure—a key requirement for large-scale systems. Using QEC to extend the qubit lifetime remains one of the outstanding experimental challenges in quantum computing. Here we report the protection of classical states from environmental bit-flip errors and demonstrate the suppression of these errors with increasing system size. We use a linear array of nine qubits, which is a natural step towards the two-dimensional surface code QEC scheme, and track errors as they occur by repeatedly performing projective quantum non-demolition parity measurements. Relative to a single physical qubit, we reduce the failure rate in retrieving an input state by a factor of 2.7 when using five of our nine qubits and by a factor of 8.5 when using all nine qubits after eight cycles. Additionally, we tomographically verify preservation of the non-classical Greenberger–Horne–Zeilinger state. The successful suppression of environment-induced errors will motivate further research into the many challenges associated with building a large-scale superconducting quantum computer.