I would like to ask why in the MRB demo code, the gate set you selected does not include all single-qubit Clifford gates. It is written in the article that this gate set must satisfy 2-design, and the complete 24 Clifford gates is a 2-design.
Additionally, why does the fit parameter r_Ω become smaller when using the Pauli randomized layer with all the one-qubit gate having the same error rate?
I would like to ask why in the MRB demo code, the gate set you selected does not include all single-qubit Clifford gates. It is written in the article that this gate set must satisfy 2-design, and the complete 24 Clifford gates is a 2-design.