Probabilistic error cancellation for measurement-based quantum computation
A method of probabilistically canceling noise in a measurement-based quantum device includes obtaining a sequence of ideal measurements included within a quantum algorithm and selecting a sequence of noisy measurements for emulating the sequence of ideal measurements. Each of the noisy measurements in the selected sequence approximates a corresponding one of the ideal measurements and is adjusted by a quantum correction, where the noisy measurements are selected according to a carefully chosen distribution to cancel known features of noise in those same noisy measurements in the sequence.
1. A method for removing noise from an expectation value of a quantum algorithm implemented by a noisy quantum circuit, the method comprising:
receiving, at a processor, a sequence of ideal quantum operations included within the quantum algorithm;
implementing each ideal quantum operation in the sequence of ideal quantum operations by a series of processing operations including:
selecting a noisy measurement instrument to emulate an operation corresponding to the ideal quantum operation;
observing a noisy outcome of a measurement operation performed using the selected noisy measurement instrument in the noisy quantum circuit;
sampling from a probability distribution to obtain an adjusted outcome for the measurement operation, the probability distribution representing a spread of possible outcomes given the observed noisy outcome and the selected noisy measurement instrument; and
determining and storing bias correction weights for the measurement operation, the bias correction weights being based on the observed noisy outcome, the selected noisy measurement instrument, and the sampled adjusted outcome; and
after each iteration of multiple iterations of the quantum algorithm in the noisy quantum circuit, removing noise from an output of the quantum algorithm by re-scaling the output by a scalar computed based on the bias correction weights.
2. The method of claim 1 , further comprising:
computing an adjusted expectation value based on re-scaled outputs from the multiple iterations of the quantum algorithm.
3. The method of claim 1 , further comprising:
based on the sampled adjusted outcome, determining a quantum correction to adjust a quantum state;
absorbing the quantum correction into a next noisy operation corresponding to a next ideal operation in the sequence of ideal operations.
4. The method of claim 3 , wherein selecting the noisy measurement instrument to emulate the operation corresponding to the ideal quantum operation comprises:
selecting the noisy instrument to emulate a product of the next ideal operation and the absorbed quantum correction.
5. The method of claim 3 , wherein the observed noisy outcome is a soft outcome and wherein sampling from the probability distribution is based on the soft outcome.
6. The method of claim 3 , wherein determining the quantum correction further comprises:
retrieving the quantum correction from a table, the quantum correction being stored in association with the selected noisy measurement instrument and the observed noisy outcome.
7. The method of claim 1 , wherein the adjusted expectation value matches an expectation value of the sequence of ideal quantum operations.
8. The method of claim 1 , wherein the method is effective to cancel select noise characteristics from the expectation value of the quantum algorithm, the select noise characteristics comprising less than a total quantity of noise present in the noisy quantum circuit.
9. The method of claim 8 , wherein the select noise characteristics include outcome errors of the noisy quantum circuit.
10. A quantum computing system comprising:
a controller including at least a processor and processor-executable instructions, the controller configured to:
receive a sequence of ideal quantum operations included within a quantum algorithm;
implement each ideal quantum operation in the sequence of ideal quantum operations by a series of processing operations including:
selecting a noisy measurement instrument to emulate an operation corresponding to the ideal quantum operation;
observing a noisy outcome of a measurement operation performed using the selected noisy measurement instrument in a noisy quantum circuit;
sampling from a probability distribution to obtain an adjusted outcome for the measurement operation, the probability distribution representing a spread of possible outcomes given the observed noisy outcome and the selected noisy measurement instrument; and
determining and storing bias correction weights for the measurement operation, the bias correction weights being based on the observed noisy outcome, the selected noisy measurement instrument, and the sampled adjusted outcome; and
after each iteration of multiple iterations of the quantum algorithm in the noisy quantum circuit, removing noise from an output of the quantum algorithm by re-scaling the output by a scalar computed based on the bias correction weights.
11. The quantum computing system of claim 10 , wherein the controller is further configured to compute an adjusted expectation value based on re-scaled outputs from the multiple iterations of the quantum algorithm.
12. The quantum computing system of claim 11 , wherein the adjusted expectation value matches an expectation value of the sequence of ideal quantum operations.
13. The quantum computing system of claim 10 , wherein the controller is further configured to:
based on the sampled adjusted outcome, determine a quantum correction to adjust a quantum state, wherein implementing a next ideal operation in the series of ideal operations further includes selecting a noisy instrument to emulate a product of the next ideal operation and the quantum correction.
14. The quantum computing system of claim 13 , wherein determining the quantum correction further comprises:
retrieving the quantum correction from a table, the quantum correction being stored in association with the selected noisy measurement instrument and the observed noisy outcome.
15. The quantum computing system of claim 10 , wherein the observed noisy outcome is a soft outcome and wherein sampling from the probability distribution is based on the soft outcome.
16. One or more non-transitory computer-readable storage media storing instructions for executing a computer process, the computer process comprising:
receiving a sequence of ideal quantum operations included within a quantum algorithm;
implementing each ideal quantum operation in the sequence of ideal quantum operations by a sequence of operations that includes:
selecting a noisy measurement instrument to emulate an operation corresponding to the ideal quantum operation;
observing a noisy outcome of a measurement operation performed using the selected noisy measurement instrument in a noisy quantum circuit;
sampling from a probability distribution to obtain an adjusted outcome for the measurement operation, the probability distribution representing a spread of possible outcomes given the observed noisy outcome and the selected noisy measurement instrument; and
determining and storing bias correction weights for the measurement operation, the bias correction weights being based on the observed noisy outcome, the selected noisy measurement instrument, and the sampled adjusted outcome;
after each iteration of multiple iterations of the quantum algorithm in the noisy quantum circuit, removing noise from an output of the quantum algorithm by re-scaling the output by a scalar computed based on the bias correction weights.
17. The one or more non-transitory computer-readable storage media of claim 16 , wherein removing the noise from the output of the quantum algorithm for each of the multiple iterations adjusts an expectation value of the quantum algorithm to match an expectation value of the sequence of ideal quantum operations.