Quantum error correction (QEC) protects quantum information from noise by encoding logical states into a code subspace or subsystem. Beyond storage, quantum computation also requires controlled processing of encoded logical information. This is accomplished by applying transforms to logical states while preserving the code. These circuits must be fault-tolerant. Errors must not spread faster than the code’s ability to correct them, or, for purely error-detecting (distance-2) codes, at least to detect them.

Code-switching offers a route to universal, fault-tolerant quantum computation by circumventing the limitation implied by the Eastin-Knill theorem against a universal transversal gate set within a single quantum code. Here, the researchers present a fault-tolerant code-switching protocol between two versions of the [[8,3,2]] code. One version supports weakly fault-tolerant single-qubit Clifford gates, while the other supports a logical CCZ gate via transversal T/T† together with logical CZ, CNOT, and SWAP gates.

Because both codes have distance 2, the protocol operates in a postselected, error-detecting regime: single faults lead to detectable outcomes, and accepted runs exhibit quadratic suppression of logical error rates. This yields a universal scheme for postselected fault-tolerant computation. We validate the protocol numerically through simulations of state preparation, code switching and a three-logical-qubit implementation of Grover’s search.

Code Switching Protocol via Gauge Fixing Between Code Versions

Fixing the gauge qubits of the [[8,3,2]] code into the logical |0⟩ state generates a version of the code supporting a logical CCZ gate via transversal T and $T^{}$ † gates, alongside CZ, CNOT, and SWAP gates, a construction important to enabling universal quantum computation through code switching. This version, designated Version 2 of the [[8,3,2]] code, differs from its counterpart, Version 1, which instead facilitates weakly fault-tolerant single-qubit gates like H, S, and, achieved through specific XX and ZZ rotation gadgets.

The ability to move between these two code versions, while maintaining logical information, relies on a process called gauge fixing, involving the measurement of gauge operators and projection onto eigenstates that become stabilizer generators of the target code.

The development of this protocol circumvents limitations imposed by the Eastin-Knill theorem, a long-standing challenge in quantum computing, by offering a route to universal fault-tolerant computation without requiring a universal set of transversal gates within a single code. This is achieved by using the complementary gate sets available in each version of the [[8,3,2]] code, effectively creating an overcomplete universal set when combined.

The researchers demonstrated that single inter-block logical CNOT gates are available between blocks using the same code version, but only operate in one direction, from Version 2 to Version 1, when switching between versions. This directional constraint is a key characteristic of the protocol and influences the design of using code switching. Simulations of state preparation, code switching and a three-logical-qubit implementation of Grover’s search algorithm validate the protocol’s efficacy, revealing quadratic suppression of both postselected logical error rates and decoded-output failure probability.

The Grover’s algorithm benchmark is particularly significant because its logical circuit requires both a logical H gate and a logical CCZ gate, necessitating the switching between the two code versions to implement it fully. The foundation of this approach lies in the [[8,3,2]] subsystem code, derived from the [[n,n-2,2]] stabilizer code by “gauging out” three of its six logical operators, effectively removing them from data storage and protection.

This process involves a noncanonical choice of logical basis, important for the code-switching protocol to function correctly. The researchers explain that fixing the gauge qubits in the | + ⟩ state yields Version 1, while fixing them in the |0⟩ state yields Version 2, demonstrating that both versions originate from the same underlying subsystem code. This connection suggests a broader family of codes, potentially extending the principles of code switching to higher-dimensional hypercube codes.

The researchers note that the [[8,3,2]] code belongs to the [[ $2^{D}, D, 2$ ]] hypercube family with D = 3, while the parent code belongs to the [[n,n-2,2]] family. Their use of a noncanonical logical basis raises the question of how to systematically choose logical bases for [[ $2^{D}, 2^{D} - 2, 2$ ]] codes, allowing $2^{D} - 2 - D$ logical qubits to be gauged out while preserving useful transversal structure.

The paper clarifies, detailing the specific gates supported by each version. “Fixing them in | + _g⟩ yields Version 1, in which the resulting weight-2 representatives of the logical operators enable weakly fault-tolerant implementations of H, S, and The work’s conclusion emphasizes the development of a fault-tolerant, error-detecting protocol for switching between the two versions of the [[8,3,2]] code, enabling the assembly of a universal logical gate set.

The researchers suggest that their construction hints at a broader family viewpoint, potentially generalizing the Version 1 construction and opening avenues for further exploration of code switching techniques. This suggests that the principles demonstrated with the [[8,3,2]] code could be extended to other codes within these families, potentially leading to more efficient and versatile quantum computation schemes. The simulations consistently demonstrate quadratic suppression of postselected logical error rates, confirming the protocol’s ability to achieve weak fault tolerance.

Grover’s Search Implemented with Three Logical Qubits

This validation extended to a three-logical-qubit implementation of Grover’s search. The ability to implement these gates within Version 2 expands the computational possibilities of the [[8,3,2]] code, providing a versatile platform for complex computations. Logical SWAP gates, important for qubit rearrangement, are induced by physical qubit permutations, potentially using dynamically reconfigurable hardware architectures like neutral-atom arrays or shuttling trapped-ion systems to absorb these operations into software-level qubit remapping.

Scaling this approach beyond individual code blocks requires entangling gates between logical qubits residing in different blocks. A parallel inter-block logical CNOT gate, for example, applies transversal physical CNOT gates between corresponding qubits, enabling parallel operations on multiple logical qubits. This demonstrates a pathway to building larger, more complex systems by interconnecting smaller, manageable code blocks. The code-switching protocol itself relies on transitioning between these two versions of the [[8,3,2]] code, a process that requires careful consideration of gauge measurements.

Switching from Version 1 to Version 2 necessitates flagged measurements of weight-4 Pauli-Z gauge operators, while the reverse direction uses weight-2 Pauli-X gauge measurements, requiring fewer flag qubits. Shixin Wu, Dawei Zhong, Todd A. Brun, Daniel A. Lidar validated the protocol through numerical simulations, focusing on the preservation of logical states during the switching process. A benchmark specifically designed to assess this preservation involved a state-specific diagnostic, using the properties of the | + + + ⟩ state, where pure logical X-type faults act trivially.

The simulations consistently demonstrated quadratic suppression of the postselected decoded-output failure probability, aligning with the expectation of weak fault tolerance. The underlying principle of this work extends beyond the specific [[8,3,2]] code, hinting at a broader family of codes and potentially generalizing the code-switching approach. This opens up possibilities for exploring alternative error-correcting codes and developing more robust and scalable quantum computing architectures. The authors state, “This yields a universal scheme for postselected fault-tolerant computation,” emphasizing the potential for building practical quantum computers that can overcome the limitations of current error-correction techniques.

Eastin-Knill Theorem and the Need for Code Switching Approaches

The limitations of the Eastin-Knill theorem, a longstanding obstacle to universal quantum computation, are addressed by a new protocol employing code switching between two distinct versions of the [[8,3,2]] code, as demonstrated through numerical simulations. Central to this work is the concept of gauge fixing, where measurements of specific operators project the quantum state onto a desired basis within the target code, ensuring compatibility before subsequent logical operations are applied. This technique, combined with error detection, allows the system to identify and discard faulty runs, maintaining a high level of fidelity in the accepted results.

The simulations reveal that lower acceptance rates, indicating more frequent discarding of erroneous runs, do not necessarily equate to lower fidelity of the remaining states. Rather, they reflect a more stringent filtering process that prioritizes accuracy. This selective acceptance is important for achieving weak fault tolerance, where errors are suppressed but not fully corrected, relying instead on the probability of detecting and discarding faulty outcomes. This suppression is a theoretical prediction and a numerically verified outcome, bolstering confidence in the feasibility of this approach.

👉 More information
🗞 Universal weakly fault-tolerant quantum computation via code switching in the [[8,3,2]] code
✍️ Shixin Wu, Dawei Zhong, Todd A. Brun and Daniel A. Lidar
🧠 DOI: http://link.aps.org/doi/10.1103/4gwn-6cnx

More like this