To improve the efficiency of working with Pauli error channels in operators and noise models, particularly with a focus of large number of qubit simulations on stabilizer or MPS simulators.
I am working on 2 specific changes to Aer at the moment:
[x] #2157 Add a PauliLindbladError noise error class supported by Aer's noise model and C++ sampling to represent Pauli channels generated by Pauli lindblad dissipators. When these are sparse generators these can be sampled from efficiently for large numbers of qubit errors like is used in PEC and PEA mitigation experiments.
[x] #2156 Add a PauliError noise error class to more efficiently work with Pauli channels as operators (for composition, tensor product, etc) than the current QuantumError format which represents Pauli channels as QuantumCircuits. This new operator will use a PauliTable internally, but still serialize to the same structure as a pauli channel QuantumError for running on C++ simulators so that they don't need any additional changes.
What is the expected behavior?
To improve the efficiency of working with Pauli error channels in operators and noise models, particularly with a focus of large number of qubit simulations on stabilizer or MPS simulators.
I am working on 2 specific changes to Aer at the moment:
PauliLindbladErrornoise error class supported by Aer's noise model and C++ sampling to represent Pauli channels generated by Pauli lindblad dissipators. When these are sparse generators these can be sampled from efficiently for large numbers of qubit errors like is used in PEC and PEA mitigation experiments.PauliErrornoise error class to more efficiently work with Pauli channels as operators (for composition, tensor product, etc) than the current QuantumError format which represents Pauli channels as QuantumCircuits. This new operator will use a PauliTable internally, but still serialize to the same structure as a pauli channel QuantumError for running on C++ simulators so that they don't need any additional changes.