Qudit non-Clifford interleaved benchmarking

David Amaro-Alcalá,Barry C. Sanders,Hubert de Guise

DOI: https://doi.org/10.1109/ISMVL60454.2024.00029

2024-08-26

Abstract:We introduce a scheme to characterise a qudit T gate that has different noise than a set of Clifford gates. We developed our scheme through representation theory and ring theory to generalise non-Clifford interleaved benchmarking to qudit systems. By restricting to the qubit case, we recover the dihedral benchmarking scheme. Our characterisation scheme provides experimental physicists a practical method for characterising universal qudit gate sets and advances randomised benchmarking research by providing the characterisation of a complete qudit library.

Quantum Physics

What problem does this paper attempt to address?

This paper aims to solve the individual characterization problem of non - Clifford gates in multi - level systems (qudit) in quantum computing. Specifically, the authors propose a new method to estimate the average gate fidelity of non - Clifford qudit gates with different noise characteristics. This method is based on the randomized benchmarking (RB) technique and extends it to the qudit system through representation theory and ring theory tools, thus providing a practical method to characterize the universal qudit gate set and promoting the development of randomized benchmarking research.

### Main Contributions
1. **Method Innovation**: A new scheme is introduced to characterize the T - gates of the qudit system, which have noise characteristics different from those of Clifford gates and Pauli gates.
2. **Theoretical Basis**: Using representation theory and ring theory tools, the non - Clifford interleaved benchmarking (IB) method is generalized to the qudit system.
3. **Practical Value**: It provides experimental physicists with a practical method to characterize the universal qudit gate set, especially in multi - level quantum computing systems.
4. **Technical Details**:
- **Circuit Design**: A circuit is designed that can effectively combine the noise of T - gates and Clifford - like gates, thereby estimating the average gate fidelity of the composite noise.
- **Parameter Estimation**: Key parameters are estimated through the sequence fidelity measurement of specific states, and then the average gate fidelity is calculated.
- **Numerical Verification**: The effectiveness of the method is verified by numerical simulation, and an application example in the qutrit system is shown.

### Background
- **Quantum Computing**: Multi - level quantum systems (qudit) have potential advantages in quantum computing, but the characterization of their gates remains a challenge.
- **Randomized Benchmarking**: RB is a widely used method for quantum gate characterization, but the traditional RB method is mainly applicable to qubit systems, and there is still a lack of effective solutions for the non - Clifford gate characterization of qudit systems.

### Method Overview
1. **Constructing Clifford - like Subgroup**: A Clifford - like subgroup is constructed by generating a subgroup consisting of Clifford gates and cyclic mappings.
2. **Interleaved Benchmarking**: A circuit is designed to apply T - gates and Clifford - like gates interleavedly, and the average gate fidelity of T - gates is estimated by measuring the sequence fidelity.
3. **Parameter Estimation**: Key parameters are estimated through the sequence fidelity measurement of specific states, and then the average gate fidelity is calculated.
4. **Numerical Simulation**: The effectiveness of the method is verified by numerical simulation, and an application example in the qutrit system is shown.

### Conclusion
This paper proposes a new method that can effectively characterize the non - Clifford gates of the qudit system, especially the T - gates. This method is not only of great theoretical significance but also provides a practical tool for gate characterization in multi - level quantum computing systems in practical applications.