Adaptive basis selection for fusion measurements

Patent metadata

Field	Value
Publication number	US-11367013-B2
Application number	US-202117216428-A
Country	US
Kind code	B2
Filing date	Mar 29, 2021
Priority date	Mar 30, 2020
Publication date	Jun 21, 2022
Grant date	Jun 21, 2022

Abstract

A quantum computing system and methods for performing fault-tolerant quantum computing. A fusion controller sequentially performs a series of fusion measurements on different fusion sites of a plurality of fusion sites to obtain a respective series of classical measurement results. The series of fusion measurements is performed on quantum modes of a logical qubit. For respective fusion measurements of the series of fusion measurements, a basis for performing the respective fusion measurement is selected based on classical measurement results of previous fusion measurements. The series of classical measurement results are in the memory medium.

First claim

What is claimed is:

1. A method comprising:
performing a first fusion measurement on a first quantum mode and a second quantum mode at a first fusion site, wherein the first fusion measurement comprises a two-qubit projective entanglement measurement of the first and second quantum modes;
obtaining a first classical measurement result of the first fusion measurement;
selecting a basis for performing a second fusion measurement based at least in part on the first classical measurement result;
performing the second fusion measurement on a third quantum mode and a fourth quantum mode at a second fusion site according to the selected basis, wherein the second fusion measurement comprises a two-qubit projective entanglement measurement of the third and fourth quantum modes; and
obtaining a second classical measurement result of the second fusion measurement; and
storing the first and second classical measurement results in a non-transitory memory medium.

2. The method of claim 1 ,
wherein selecting the basis for performing the second fusion measurement comprises determining whether to apply a Hadamard gate at the second fusion site while performing the second fusion measurement.

3. The method of claim 1 , the method further comprising:
sequentially performing one or more third fusion measurements at one or more respective third fusion sites;
selecting bases for performing each of the one or more third fusion measurements based at least in part on classical measurement results of previous fusion measurements;
obtaining third classical measurement results of the one or more third fusion measurements; and
storing the third classical measurement results in the non-transitory memory medium.

4. The method of claim 1 , the method further comprising:
computing an output of a quantum computational algorithm based at least in part on the first and second classical measurement results; and
storing the output in the non-transitory memory medium.

5. The method of claim 1 ,
wherein the first, second, third and fourth quantum modes are part of a logical qubit, and
wherein the basis for performing the second fusion measurement is selected to reduce the likelihood of a logical error in the logical qubit.

6. The method of claim 1 ,
wherein the first, second, third and fourth quantum modes are part of a logical qubit, and
wherein the basis for performing the second fusion measurement is selected from a first basis that risks a first parity check error in a primal syndrome graph of the logical qubit and a second basis that risks a second parity check error in a dual syndrome graph of the logical qubit.

7. The method of claim 1 ,
wherein the first, second, third and fourth quantum modes are part of a logical qubit described by a syndrome graph,
wherein the syndrome graph is updated based on the first and second classical measurement results, and
wherein selecting the basis for performing the second fusion measurement comprises:
analyzing the syndrome graph to identify a connected component of the syndrome graph, wherein the connected component comprises a contiguous set of erased edges in the syndrome graph; and
determining that a first basis risks an erasure of an edge of the syndrome graph that connects two nodes within the connected component, wherein the first basis is the selected basis for performing the second fusion measurement.

8. The method of claim 7 , further comprising:
disabling boosting for the second fusion measurement based at least in part on determining that the first basis risks the parity check error connecting two nodes within the connected component.

9. The method of claim 1 ,
wherein the first, second, third and fourth quantum modes are part of a logical qubit described by a syndrome graph, wherein the syndrome graph comprises a primal syndrome graph and a dual syndrome graph,
wherein selecting the basis for performing the second fusion measurement comprises:
analyzing the primal syndrome graph to identify a first connected component and a second connected component, wherein the first and second connected components each comprise a contiguous set of erased edges of the primal syndrome graph, wherein performing the second measurement in a first basis risks a first parity check error that connects the first and second connected components;
analyzing the dual syndrome graph to identify a third connected component and a fourth connected component, wherein the third and fourth connected components each comprise a contiguous set of erased edges of the dual syndrome graph, wherein performing the second measurement in a second basis risks a second parity check error that connects the third and fourth connected components;
comparing a first product of exposures of the first and second connected components to a second product of the exposures of the third and fourth connected components to determine which of the first or second products is a smaller product, wherein the exposures each comprise a sum of a number of edges connected to the respective connected component that have not been measured;
selecting the basis for performing the second fusion measurement to be a basis corresponding to the smaller product, wherein the first basis corresponds to the first product and the second basis corresponds to the second product.

10. The method of claim 1 ,
wherein the first, second, third and fourth quantum modes are part of a logical qubit described by a syndrome graph, wherein the syndrome graph comprises a primal syndrome graph and a dual syndrome graph,
wherein selecting the basis for performing the second measurement comprises:
calculating an exposure of each of two connected components for each of the primal and dual syndrome graphs, wherein each connected component comprises a set of contiguous erased edges of the primal or dual syndrome graph, wherein the exposures comprise a sum of a number of unmeasured edges connected to the respective connected component; and
selecting the basis from a first basis that risks erasing an edge that connects the two connected components of the primal syndrome graph and a second basis that risks erasing an edge that connects the two connected components of the dual syndrome graph based at least in part on a comparison of a first product of the exposures of the two connected components of the primal syndrome graph and a second product of the exposures of the two connected components of the dual syndrome graph.

11. The method of claim 10 ,
wherein the sum comprises a weighted sum that computes a second moment of the unmeasured edges around their mean position in the primal or dual syndrome graph.

12. The method of claim 10 , the method further comprising:
determining that the first product differs in magnitude from the second product by more than a predetermined threshold amount; and
based at least in part on determining that the first product differs in magnitude from the second product by more than the predetermined threshold amount, enabling boosting for the second fusion measurement.

13. The method of claim 10 ,
wherein the sum further comprises a weighted sum of a number of measured edges connected to the respective connected component, wherein the measured edges have a smaller weighting in the sum than the unmeasured edges.

14. The method of claim 10 ,
wherein selecting the basis from the first basis and the second basis based at least in part on a comparison of the first and second products of the exposures comprises calculating a parameter β based on the first and second products of the exposures and at least one numerical parameter and determining a

Assignees

• Psiquantum Corp

Inventors

• Pant Mihir

Classifications

• G06N10/40

  Physical realisations or architectures of quantum processors or components for manipulating qubits, e.g. qubit coupling or qubit control · CPC title

• G06N10/70

  Quantum error correction, detection or prevention, e.g. surface codes or magic state distillation · CPC title

• G06F11/25 Primary

  Testing of logic operation, e.g. by logic analysers · CPC title

• G06N10/60

  Quantum algorithms, e.g. based on quantum optimisation, quantum Fourier or Hadamard transforms · CPC title

• G06N10/80

  Quantum programming, e.g. interfaces, languages or software-development kits for creating or handling programs capable of running on quantum computers; Platforms for simulating or accessing quantum computers, e.g. cloud-based quantum computing · CPC title