Systems and methods for reducing errors in calibrated devices
Methods for reducing errors in calibrated devices comprise detecting outliers, self-checking consistency of measurements, tuning device controls to target values, and absolutely calibrating devices via a first standard and cross-checking the results via a second standard. The first standard may be a calibrated current and the second calibration standard may be a calibrated frequency. A calibrated frequency may be a microwave signal applied to the body of a qubit. Qubit annealing controls can quickly lower and raise the tunnel barrier to measures the oscillation frequency of the qubit between two potential wells.
1. A method for reducing errors in a processor system comprising a first processor and a second processor, the second processor comprising a plurality of devices, each device in the plurality of devices communicatively coupled to at least one other device via a coupler, each coupler having at least one determinable parameter, and a programming interface communicatively coupled to each device of the plurality of devices, the method comprising:
iteratively repeating until a respective first and second value of the least one determinable parameter of every coupler has been obtained:
selecting an i th device from the plurality of devices of the second processor via the first processor;
identifying all couplers communicatively coupled to the i th device via the first processor;
applying a calibration signal to the i th device via the first processor;
obtaining a respective value of the at least one determinable parameter of every coupler communicatively coupled to the i th device; and
using the first and second value of the at least one determinable parameter to determine if any of the devices in the plurality of devices are outlier devices via the first processor.
2. The method of claim 1 wherein:
selecting an i th device from the plurality of devices of the second processor via the first processor includes selecting an i th qubit of a quantum processor via a digital processor;
identifying all couplers communicatively coupled to the i th device via the first processor includes identifying all couplers inductively coupled to the i th qubit via the digital processor;
applying a calibration signal to the i th device via the first processor includes applying a calibration signal to the i th qubit via the digital processor;
obtaining a respective value of the at least one determinable parameter of every coupler communicatively coupled to the i th device includes obtaining a respective value of a mutual inductance of every coupler inductively coupled to the i th qubit; and
using the first and second value of the at least one determinable parameter to determine if any of the devices are outlier devices via the first processor includes using the first and second value of the mutual inductance to determine which of the devices in the plurality of devices are outlier qubits, if any of the devices in the plurality of devices are outlier devices, via the digital processor.
3. The method of claim 2 , further comprising programming the quantum processor to disregard outlier qubits, if any of the devices are outlier qubits.
4. A method for reducing errors in a processor system comprising a first processor and a second processor, the second processor comprising a plurality of devices, each device in the plurality of devices having at least one determinable parameter, and a programming interface communicatively coupled to each device in the plurality of devices, the method comprising:
applying a calibration signal to a first device via the first processor;
reading out a first device first value of a determinable parameter of the first device, wherein the first device first value depends on the calibration signal applied to the first device;
applying the calibration signal to a second device via the first processor, wherein the second device is communicatively coupled to the first device;
reading out a second device first value of the determinable parameter of the second device, wherein the second device first value depends on the calibration signal applied to the second device;
reading out a first device second value of the determinable parameter of the first device, wherein the first device second value depends on the calibration signal applied to the second device;
applying the calibration signal to the first device via the first processor;
reading out a second device second value of the determinable parameter of the second device, wherein the second device second value depends on the calibration signal applied to the first device;
calculating a first device first measure from the first device first value and the first device second values of the determinable parameter of the first device;
calculating a second device first measure from the second device first value and the second device second values of the determinable parameter of the second device; and
programming the first and the second device via the programming interface using the first device first measure and the second device first measure of the determinable parameter of the first and second device.
5. The method of claim 4 wherein:
applying a calibration signal to a first device via the first processor includes applying a calibration signal to a first qubit via a digital processor;
reading out a first device first value of a determinable parameter of the first device, wherein the first device first value depends on the calibration signal applied to the first device includes reading out a first device first value of a determinable parameter of the first qubit, wherein the first device first value depends on the calibration signal applied to the first qubit;
applying the calibration signal to a second device via the first processor, wherein the second device is communicatively coupled to the first device includes applying the calibration signal to a second qubit via the digital processor, wherein the second qubit is communicatively coupled to the first qubit;
reading out a second device first value of the determinable parameter of the second device, wherein the second device first value depends on the calibration signal applied to the second device includes reading out a second device first value of the determinable parameter of the second qubit, wherein the second device first value depends on the calibration signal applied to the second qubit;
reading out a first device second value of the determinable parameter of the first device, wherein the first device second value depends on the calibration signal applied to the second device includes reading out a first device second value of the determinable parameter of the first qubit, wherein the first device second value depends on the calibration signal applied to the second qubit;
applying the calibration signal to the first device via the first processor includes applying the calibration signal to the first qubit via the first processor;
reading out a second device second value of the determinable parameter of the second device, wherein the second device second value depends on the calibration signal applied to the first device includes reading out a second device second value of the determinable parameter of the second qubit, wherein the second value depends on the calibration signal applied to the first qubit;
calculating a first device first measure from the first device first value and the first device second values of the determinable parameter of the first device includes averaging the first device first value and the second device second values of the determinable parameter of the first qubit to obtain an averaged value of the determinable parameter of the first qubit;
calculating a second device first measure from the second device first value and the second device second values of the determinable parameter of the second device includes averaging the second device first and the second device second values of the determinable parameter of the second qubit to obtain an averaged value of the determinable parameter of the second qubit; and
programming the first and the second device via the programming interface using the first device first measure and the second device first measure of the determinable parameter of the first and second device includes programming the first and the second qubit via the programming interface using the respective averaged values of the determinable parameter of the first and second qubit.
6. The method of claim 5 , wherein applying a calibration signal to a first qubit includes applying a calibration signal to a first qubit via a calibration signal source, the calibration signal source electrically coupled to signal generation and control electronics external to the second processor; and applying the calibration signal to a second qubit includes applying the calibration signal to a second qubit via the calibration signal source.
7. The method of claim 5 wherein:
reading out a first device first value of a determinable parameter of the first qubit includes measuring a first device first value of a determinable parameter of the first qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing;
reading out a second device first value of the determinable parameter of the second qubit includes measuring a second device first value of the determinable parameter of the second qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing;
reading out a first device second value of the determinable parameter of the first qubit includes measuring a first device second value of the determinable parameter of the first qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing;
reading out a second device second value of the determinable parameter of the second qubit includes measuring a second device second value of the determinable parameter of the second qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing;
averaging the first device first value and the first device second value of the determinable parameter of the first qubit to obtain an averaged first device value of the determinable parameter of the first qubit includes averaging the first device first value and the first device second value of the determinable parameter of the first qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing to obtain an averaged first device value of the determinable parameter of the first qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing;
averaging the second device first value and the second device second value of the determinable parameter of the second qubit to obtain an averaged second device value of the determinable parameter of the second qubit includes averaging the second device first value and the second device second value of the determinable parameter of the second qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing to obtain an averaged second device value of the determinable parameter of the second qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing; and
programming the first and the second qubit via the programming interface using the averaged first device value and the averaged second device value of the determinable parameter of the first and second qubit includes programming the first and the second qubit via the programming interface using the averaged first device value and the averaged second device value of the determinable parameter of the first and second qubit selected from a group comprising: mutual inductance, electric moment, magnetic moment, and low energy quantum eigenspectrum spacing.
8. A method for reducing errors in a processor system comprising a first processor and second processor, the second processor comprising a plurality of devices, each device in the plurality of devices having at least one determinable parameter, and a programming interface communicatively coupled to the plurality of devices, the method comprising:
absolute calibrating the at least one determinable parameter of at least one device in the plurality of devices using a first calibration standard via the first processor; and
cross-checking the absolute calibration of the at least one determinable parameter using a second calibration standard via the first processor.
9. The method of claim 8 wherein, absolute calibrating the at least one determinable parameter of at least one device in the plurality of devices using a first calibration standard via the first processor includes absolute calibrating the at least one determinable parameter of at least one qubit in a plurality of qubits using a first calibration standard via a digital processor.
10. The method of claim 9 , wherein absolute calibrating the at least one determinable parameter of the at least one qubit in the plurality of qubits using a first calibration standard includes absolute calibrating the at least one determinable parameter of the at least one qubit in the plurality of qubits using a calibrated current, and cross-checking the absolute calibration of the at least one determinable parameter using a second calibration standard includes cross-checking the absolute calibration of the at least one determinable parameter using a calibrated frequency.
11. The method of claim 10 wherein a cross-checking the absolute calibration of the at least one determinable parameter using a calibrated frequency includes applying a microwave signal of known frequency to a body of the at least one qubit via a microwave line inductively coupled to the body of the at least one qubit.
12. The method of claim 10 wherein cross-checking the absolute calibration of the at least one determinable parameter using a calibrated frequency includes initializing the at least one qubit in one potential well; lowering a tunnel barrier of the at least one qubit via the programming interface to measure an oscillation frequency of the at least one qubit between two potential wells; and raising the tunnel barrier of the at least one qubit via the programming interface.
13. The method of claim 12 wherein lowering the tunnel barrier of the at least one qubit via the programming interface includes lowering the tunnel barrier of the at least one qubit via qubit annealing controls and raising the tunnel barrier of the at least one qubit via the programming interface includes raising the tunnel barrier of the at least one qubit via qubit annealing controls.
14. The method of claim 13 further comprising reading out a state of the at least one qubit.