Experimental considerations for zero noise extrapolation
ORAL
Abstract
While decoherence and control errors limit the size of quantum computation in the absence of fault tolerance, a number of error mitigation techniques have been developed to access noise free estimates of expectation values. In particular, the zero-noise extrapolation technique was shown to extend the computational reach of a noisy superconducting processor. Here, quantum circuits are run at amplified noise rates, to extrapolate the results of these runs to the zero noise limit. Under the assumption of time invariant noise, such noise amplification may be achieved by stretching in time the gates employed in the circuit. We test these assumptions for superconducting processors with all-microwave drives and also consider the effect of errors in the gate rescaling on the extrapolation.
–
Presenters
-
Abhinav Kandala
IBM TJ Watson Research Center
Authors
-
Abhinav Kandala
IBM TJ Watson Research Center
-
Kristan Temme
IBM TJ Watson Research Center
-
Seth Merkel
IBM TJ Watson Research Center
-
David McKay
IBM TJ Watson Research Center
-
Easwar M Magesan
IBM TJ Watson Research Center
-
Jay M Gambetta
IBM TJ Watson Research Center