This code allows you to reproduce the experiments performed by Rosenberg et al. Quantum Science and Technology 7, 015024 (2022). Specifically, it contains functions to optimize variational ansatze both classically and hybridly, as well as functions to implement a variety of error-mitigation techniques.
Version numbers indicate the version I used. Other versions may also be compatible.
qiskit 0.23.0
qiskit-aer 0.7.0
qiskit-ibmq-provider 0.11.0
numpy 1.19.1
scipy 1.5.2
cupy 8.0.0 (optional; only if you want to use a GPU to perform classical evaluations)
qiskit-aer-gpu 0.7.4 (optional; only for qiskit aer simulations using the gpu)
First, install the dependencies.
If you want to run a Variational Quantum Eigensolver (VQE) experiment, first modify optimize_machine(...) in optimization.py so that it saves the measured energy (E) and the classically evaluated energy (E_exact) to a file of your choice at each evaluation. Then modify VQE.py to your situation and execute VQE.py.
If you want to test the various error mitigation techniques described in Rosenberg et al (2021), follow the directions given as comments in error_mitigation.py. Search this file and optimization.py for /your_directory/ and replace with the directory where you want to save data.
VQE.py can be executed to run a Variational Quantum Eigensolver (VQE) that uses measurements from actual quantum computers to tune an ansatz to the ground state of the Mixed-Field Ising Model. It should first be edited to suit your needs, and you may want to edit the cost function to save the measured energy at each evaluation. It can be easily modified to use a different Hamiltonian.
optimization.py contains functions that implement classical or hybrid optimization of variational ansatze. optimize_machine(...) is the function called in VQE.py that performs hybrid classical-quantum optimization. optimize(...) performs a classical optimization. optimize_symm(...) performs classical optimization, imposing permutation symmetry on the ansatz. See the file contents for more details.
energy_evaluation.py contains functions that submit ansatz circuits to IBM quantum computers and that extract the measured energies from the results. See the file contents for more details.
error_mitigation.py contains (as comments) human-readable instructions for benchmarking various error mitigation techniques. It also contains functions that implement these various error mitigation techniques.
library.py contains functions that are called by functions in the other files. (Although some functions in energy_evaluation.py are called in error_mitigation.py, etc.)
compiling.py contains functions for implementing zero noise extrapolation and randomized compiling.
If you use this code in your research, please consider citing the following paper:
E. Rosenberg, P. Ginsparg, and P.L. McMahon. Experimental error mitigation using linear rescaling for variational quantum eigensolving with up to 20 qubits. Quantum Science and Technology 7, 015024 (2022)
If you have any questions or comments about this code, please contact Eliott Rosenberg at enr27--at--cornell--dot--edu.
The code in this repository is released under the following license:
Creative Commons Attribution 4.0 International
A copy of this license is given in this repository as license.txt.