量子コンピューティング：キャット・コードが量子エラーを訂正し、オーバーヘッドを削減する

The demonstration of quantum error correction requiring reduced hardware overheads is reported by researchers from Amazon Web Services (AWS) in a paper published in Nature. The system uses so called ‘cat qubits’ (qubits are the quantum equivalent to classical computing bits), which are designed to be resistant against certain types of noise and errors that might disrupt the output of quantum systems. This approach requires fewer overall components to achieve quantum error correction than other designs.

Quantum computers are prone to errors, which limits their potential to exceed the capabilities of classical computers at certain tasks. Quantum error correction is a method that helps reduce errors by spreading information over multiple qubits, allowing the identification and correction of errors without corrupting the computation. However, most approaches to quantum error correction typically relies on a large number of additional qubits to provide sufficient protection against errors, potentially leading to an overall decrease in efficiency.

Harald Putterman and colleagues explore a potentially more efficient way to implement quantum error correction using a type of qubit called a bosonic cat qubit. These cat qubits are intrinsically, at the hardware level, highly resistant to one type of error (called a bit flip) at the expense of being more likely to experience another kind (called a phase flip). This error bias allows researchers to design quantum error correcting codes that focus only on dealing with phase-flip errors, resulting in an overall much more efficient design that requires fewer additional qubits. The authors demonstrate a superconducting quantum circuit device using an array of cat qubits, where errors are shown to be suppressed from 1.75% per cycle to 1.65% for an error correcting code with five cat qubits. Achieving suppression of errors with larger error-correcting code has previously required tens of additional qubits.

These results show that using bosonic cat qubits may be an efficient way to achieve fault-tolerant quantum computation. The authors suggest that the has the potential to scale, and to do so efficiently, but note that further optimization is needed to improve the performance to a level that might enable practical quantum computing applications.