Quantinuum realizes Real-Time Fault-Tolerant Quantum Error Correction for the first time
In November 2021, Honeywell Quantum Solutions (HQS) merged with Cambridge Quantum to form Quantinuum, the world's largest quantum computing company (with the largest number of employees). Currently as a subsidiary of Honeywell, Quantinuum has achieved another new achievement-an important step in proving the feasibility of large-scale quantum computing based on ion trap technology. The researchers announced that they can detect and correct quantum errors in real time. This is the first time in the industry.
The research paper has been published on PHYSICAL REVIEW X on December 23 [1], and in accordance with academic practice, Honeywell released a pre-printed version on the arXiv website in July this year. The researchers detailed how they created a single logical qubit (a series of entangled physical qubits) and applied multiple rounds of quantum error correction. This logical qubit can prevent the two main types of errors that occur in quantum computers: bit flips and phase flips.
The basis for large-scale quantum computing
Currently, most quantum error correction (QEC) demonstrations involve correcting errors or noise after the program has finished running. This technique is called post-processing. However, in order to achieve reliable large-scale quantum computing, the system must correct errors in real time.
In addition, the encoding proposed by some research groups in the past can only correct a single type of error (bit flip or phase flip, but not both at the same time), including Google, IBM/Thor and IBM/University of Basel. Furthermore, some groups have demonstrated the quantum error correction process, including Blatt and Monroe (founders of IonQ). Other groups have studied quantum error detection codes, which can detect two types of errors but cannot correct them, including ETH Zurich, Google, and Delft University of Technology.
This time, Quantinuum's method satisfies the three conditions of real-time, error detection and error correction at the same time.
The paper mentioned that in order to realize the high-level function of real-time error correction, a system is required to process several low-level primitives, including single-qubit and double-qubit operations, intermediate circuit measurements, real-time processing of measurement results, and performing these measurements. The ability to follow-up door operations.
In this work, the researchers used the color code proposed by Andrew Steane [[7, 1, 3]]. Using 10 physical qubits in Honeywell’s ion trap quantum charge coupled device (QCCD) quantum computer, a single logical qubit can be encoded, controlled and repeatedly corrected.
Top: [[7,1,3]] color code. Seven data qubits are on the vertices of the polygon, and three auxiliary qubits for syndrome measurement are on the side. Bottom: There are 10 171 Ytterbium ions (red) and 10 coolant 138 barium ions (white) in the ion trap. Ion-transport is used to arrange ions into areas for gate operation and measurement. The red and blue electrodes indicate areas that support transmission operations, including linear transmission, crystal splitting and combination, and physical exchange. The green area supports linear transmission and storage of ions between gate operations.
The logic qubits were initialized to three mutually unbiased eigenstates using an encoding circuit, and then they measured the average logical SPAM (State Preparation and Measurement) error of 1.7(2)×10−3, and the average physical SPAM error of 2.4 (8)×10−3.
Next, the researchers perform multiple syndrome measurements on the encoded qubits, using a real-time decoder to determine any necessary corrections-either as a software update to the Pauli framework or as a gate for physical applications. In addition, these processes are repeated while maintaining coherence, demonstrating a dynamically protected logical qubit memory. The above whole process is called QEC cycle (cycle).
Then, the researchers demonstrated that non-Clifford qubit operations can be performed by encoding a magic state, where the error rate needs to be lower than the threshold required for the preparation of the magic state. Finally, they proposed a system-level simulation that allowed it to identify critical hardware upgrades that allowed the system to reach a pseudo threshold.
Tony Uttley, President and Chief Operating Officer of Quantinuum, said: "Our achievements are groundbreaking. We have proved that what was once a theoretical matter is that quantum computers will be able to correct errors in real time, paving the way for precise quantum computing."
This work uses the QEC scheme first proposed by Oxford University Andrew Steane in 1996. He commented: "This is a high-quality experimental work. It is inferred by analyzing the influence of certain operations and measurements on qubits. One thing; implementation in the laboratory is another entirely. This work highlights all the parts of QEC and fault tolerance theoretical work that have proven to be of practical use."
Even Laird Egan, a quantum engineer from competitor IonQ, said, "This is a major step forward." The company has previously implemented the QEC method for ion trap qubits. "What is truly different is the multiple rounds of true fault tolerance and correction"-this is not yet achieved by the IonQ team.
Next goal
This achievement represents a milestone in large-scale quantum computing, but Quantinuum researchers are still struggling to cross the break-even point of quantum error correction-the point where the error rate of logical qubits is lower than that of physical qubits (creating logical qubits) And the application of quantum error correction codes will also bring noise to the system).
The Quantinuum team is approaching this goal. In order to cross the equilibrium point, the error rate of each QEC cycle needs to be lower than the maximum physical error rate associated with the QEC protocol.
In the paper, the researchers pointed out the key improvements needed to reach the equilibrium point. Dr. Ciaran Ryan-Anderson, Quantinuum senior physicist and the first author of the paper, said: "We believe these improvements are feasible and are pushing to complete the next step."
Researchers need to create multiple logical qubits, and this depends on advances in quantum technology, requiring better fidelity, more physical qubits, better connectivity between qubits, and other factors.
The increase of logical qubits will usher in a new era of fault-tolerant quantum computers, because even if certain operations fail, it can continue to run.
Link: [1]https://journals.aps.org/prx/abstract/10.1103/PhysRevX.11.041058
