Method for solving a quantum computer problem and quantum computer with a configurable quantum circuit

Abstract

In the method for solving a quantum computer problem, a configurable quantum circuit (QC) with an arrangement of qubits is used and a quantum computer problem (QCPROB) is received, wherein a solution algorithm is determined according to the received quantum computer problem and a configuration of the quantum circuit is determined according to the solution algorithm and a susceptibility to error of the solution algorithm, with said configuration including qubits for error correction according to the susceptibility to error of the solution algorithm, and wherein the configurable quantum circuit (QC) is configured with the determined configuration and the quantum computer problem is solved (SOL) using the configured quantum circuit (QC).

Description

[0001] Description

[0002] Method for solving a quantum computing problem and quantum computer with a configurable quantum circuit

[0003] The invention relates to a method for solving a quantum computing problem using a quantum circuit with an array of qubits and to a quantum computer with a configurable quantum circuit with an array of qubits.

[0004] In every process in electronic data processing there is the possibility of errors in the form of unwanted changes to bits. In classical computers the causes are, for example, external influences, such as cosmic radiation or radioactivity or solar flares or noise effects during computing operations or data transmission. In classical computers, however, classical bits are fault-tolerant because the states "0" and "1" or "low" and "high" cover a sufficiently large voltage range and a voltage deviation in the signal that is small compared to this voltage range therefore has little influence on the read value of a classical bit.

[0005] In quantum computers, also known as quantum computers, quantum computing operations are carried out on qubits in the conventional way instead of on classical bits. Such quantum computing operations are particularly gate operations such as NOT or CNOT and, as things stand, are highly error-prone because the energy levels for different quantum states of qubits hardly differ from one another. External influences, such as temperature caused by an electrical signal, or internal influences, such as unwanted coupling between neighboring qubits or faulty quantum gates, can cause bit flips or phase flips, i.e. changes in the phase shift, or combinations thereof. For this reason, today's quantum computers must be heavily shielded and operated at low temperatures of just a few millikelvins.

[0006] Successful computation with faulty qubits in a quantum circuit is difficult or even impossible, depending on the degree of fault tolerance of the algorithm used.

[0007] Classical error correction and error detection algorithms usually exploit redundancies in the data, such as parity bits or a multiplication of bits, so that the majority of these multiplied bits presumably indicate the intended value of the respective multiplied bits. However, such error correction and error detection algorithms are not possible with quantum computers because the no-cloning theorem prohibits the multiplication of qubit states.

[0008] Nevertheless, error correction is also possible in quantum computers. However, this requires up to 1000 physical qubits per logical qubit to be implemented for actual applications.

[0009] Even with particularly efficient quantum algorithms, error correction requires a number of additional qubits that significantly exceeds the number of logical qubits.

[0010] Current quantum computers have a very limited number of qubits. Consequently, current quantum computing problems cannot be adequately solved on such quantum computers because only a few logical qubits are available on such quantum computers.

[0011] It is therefore an object of the invention to provide a method for solving a quantum computing problem using a quantum circuit, as well as a quantum computer having a quantum circuit. This object of the invention is achieved by a method for solving a quantum computing problem using a quantum circuit, having the features specified in claim 1, and by a quantum computer having a quantum circuit, having the features specified in claim 8. Preferred developments of the invention are specified in the associated subclaims, the following description, and the drawings.

[0012] In the inventive method for solving a quantum computing problem, a configurable quantum circuit with an arrangement of qubits is used, and a quantum computing problem is received. Depending on the received quantum computing problem, a solution algorithm is determined, and depending on the solution algorithm and the error susceptibility of the solution algorithm, a configuration of the qubits is determined that has a number of qubits for error correction that depends on the error susceptibility of the solution algorithm. In the inventive method, the configurable quantum circuit is configured with the determined configuration, and the quantum computing problem is solved using the configured quantum circuit.

[0013] In the method according to the invention, a configuration of the qubits is advantageously determined depending on the solution algorithm and depending on the error susceptibility of the solution algorithm, which configuration has a number of qubits for error correction that depends on the error susceptibility of the solution algorithm. By means of this dependence on the error susceptibility of the solution algorithm, the number of qubits required for error correction can be provided for the determined solution algorithm of the respective quantum computer problem. This provided number of qubits for error correction therefore depends on the solution algorithm and thus on the respective quantum computer problem. Consequently, depending on the quantum computer problem, a number of qubits required for error correction can be provided. Consequently, only as many qubits need be provided for error correction as the error susceptibility of the solution algorithm requires.Accordingly, there is no need for a predetermined number of qubits for error correction that is independent of the specific quantum computing problem. The method according to the invention thus allows the number of qubits required for error correction to be selected as small as necessary to implement the solution algorithm. Consequently, using the method according to the invention with the quantum circuit with a predetermined number of qubits, more complex quantum computing problems can be solved than previously known. A particularly large number of the qubits of the quantum circuit can therefore be used as logical qubits, since these do not necessarily have to be available for error correction.

[0014] In the method according to the invention, the susceptibility of the solution algorithm to errors in individual steps of the solution algorithm is preferably determined, and the configuration of the qubits is determined as a function of the susceptibility of the solution algorithm to errors in the individual steps. Thus, the number of qubits for error correction, which depends on the susceptibility of the solution algorithm to errors, can be specifically tailored to the susceptibility of the solution algorithm to errors in individual steps of the solution algorithm. Preferably, the number of qubits for error correction is weighted differently for individual steps of the solution algorithm and adapted to the susceptibility of the solution algorithm to errors in these individual steps. In particular, a different number of qubits for error correction is provided for specific individual steps of the solution algorithm than for other individual steps.

[0015] In a further development of the method according to the invention, the configuration for different logical qubits for implementing the solution algorithm preferably comprises a different number of physical qubits. In this further development, physical qubits are advantageously provided as needed, depending on the relevance of errors that occur for individual logical qubits. Thus, for individual logical qubits, error correction is crucial for solving the quantum computing problem, while for other logical qubits, error correction is unnecessary. This circumstance is advantageously taken into account in this further development.

[0016] In the method according to the invention, a configuration is expediently determined that fully utilizes the available number of qubits of the quantum circuit. In this way, the configuration can be selected such that as many logical qubits as possible can be realized with the quantum circuit. Consequently, the quantum circuit can be configured such that the most complex quantum computing problem can be solved with the quantum circuit.

[0017] Advantageously, in the method according to the invention, a measure of the error-freedom of a solution determined using the solution algorithm is specified, and a configuration is determined which takes into account or fulfills the measure of the error-freedom of the solution. In this development of the invention, the measure of the error-freedom of the solution to the quantum computer problem can be used to determine the number of qubits for error correction. In particular, in cases in which only a low error-freedom of a solution determined using the solution algorithm is specified, the number of qubits for error correction can be set lower than in cases in which a higher error-freedom is specified.

[0018] In an advantageous development of the method according to the invention, a measure of the complexity of the quantum computing problem is specified and the configuration is determined depending on the measure of complexity. In this development of the invention, a complexity of the quantum computing problem required for a real problem can be specified as a boundary condition and, depending on this boundary condition, the number of qubits for error correction can be determined. In this development of the invention, the number of logical qubits for implementing the solution algorithm can therefore be specified and the error-free nature of a solution determined using the solution algorithm can be optimized, i.e. chosen to be as high as possible.

[0019] The quantum computer according to the invention has a configurable quantum circuit with an arrangement of qubits, wherein the quantum computer has an interface for receiving a quantum computing problem and wherein the quantum computer has a, preferably classical, computing device for configuring the configurable arrangement of qubits, which computing device is set up to determine a solution algorithm depending on the received quantum computing problem and to determine, depending on the solution algorithm and an error susceptibility of the solution algorithm, such a configuration of the configurable quantum circuit which has a number of qubits for error correction depending on the error susceptibility of the solution algorithm, wherein the quantum computer is set up to configure the quantum circuit according to the determined configuration.Particularly preferably, the quantum computer according to the invention is set up and designed to carry out a method according to the invention as described above.

[0020] In a preferred embodiment of the invention, the quantum computer according to the invention is designed as a single unit, i.e., as a single integrated component. Alternatively, and likewise preferably, the quantum computer according to the invention forms a multi-unit system comprising at least the quantum circuit and the computing device, which form separate components that communicate with one another.

[0021] The inventive method can be carried out using the quantum computer according to the invention. The same advantages arise for the inventive quantum computer as already explained for the inventive method. The advantageous developments of the inventive method, as described above, are preferably implemented analogously in the inventive quantum computer, i.e., the inventive quantum computer is suitably designed and configured specifically for the respective developments of the inventive method.

[0022] In the method according to the invention or in the quantum computer according to the invention, the quantum computing problem is preferably a manufacturing problem and / or a maintenance problem and / or a logistics problem and / or a medical image recognition problem. Especially in the application cases of digital manufacturing, maintenance, and logistics, complex problems regularly arise that can be solved using quantum computers and methods for solving a quantum computing problem. Therefore, the use of the method according to the invention and / or the quantum computer according to the invention is particularly advantageous in these application cases.

[0023] The invention is explained in more detail below using an exemplary embodiment illustrated in the drawing. The sole drawing figure 1 schematically shows the sequence of an inventive method for solving a quantum computing problem using an inventive quantum computer with a configurable quantum circuit in a flowchart.

[0024] The flowchart shown in Fig. 1 shows the sequence of an inventive method for solving a quantum computing problem QCPROB in the form of a digital manufacturing task on an inventive quantum computer. The digital manufacturing task concerns the efficient planning of a production process under predetermined boundary conditions. In other embodiments, the quantum computing problem QCPROB is a different problem, such as a maintenance problem or a logistics problem or a medical image recognition problem. A number of quantum algorithms for execution on a quantum computer are known for this quantum computing problem QCPROB. The quantum computing problem QCPROB is transferred to the inventive quantum computer via an interface of the quantum computer.

[0025] Using the method according to the invention, in a first step, the quantum computing problem QCPROB is received by a classical analysis computer ANA of the quantum computer, and a quantum algorithm with the smallest possible number of required logical qubits is selected to solve the quantum computing problem QCPROB. For this purpose, the analysis computer ANA has an appropriately programmed classical computing device with a database of known quantum algorithms. The selected quantum algorithm is implemented and solved on the configurable quantum circuit QC of the quantum computer in the final steps of the method.

[0026] To this end, the method first provides a logical ideal configuration of qubits, which, in the case of error-free qubits, would solve the quantum computing problem QCPROB by implementing the selected quantum algorithm. However, this logical ideal configuration of qubits underestimates the number of qubits actually required. A realistic implementation requires a large number of error-correcting qubits, which usually significantly exceeds the number of qubits for the ideal configuration—depending on the actual realization of the qubits, for example, by several orders of magnitude.

[0027] An optimization algorithm is now applied using the analysis computer ANA to calculate the proportion of error correction qubits required for a reliable solution to the quantum computing problem QCPROB. To do this, the ideal configuration of qubits is first used, and the qubits of the ideal configuration are weighted, i.e., it is determined for which qubits error correction is crucial and for which error correction is dispensable. In this way, a certain proportion of error correction qubits can be determined that is actually required for a real implementation of the ideal configuration of qubits. The qubits provided for the ideal configuration and the error correction qubits can therefore be provided in the appropriate ratio for the configuration of the quantum circuit QC.In particular, in this way it is possible to determine how many qubits are actually available in the configurable quantum circuit QC and consequently the complexity of the ideal configuration can be chosen appropriately according to the ratio of qubits for the ideal configuration and error correction qubits, so that the largest possible number of qubits is available for the ideal configuration and a sufficiently high number of error correction qubits is provided.

[0028] According to the previously performed weighting, the configuration of the configurable quantum circuit QC, comprising the ideal configuration and the specific proportion of error-correcting qubits, is configured using a configuration step OPTCONF. For this purpose, the analysis computer ANA outputs the configuration via an output interface and transfers the configuration to the quantum circuit. The quantum computer is configured to configure the quantum circuit using the configuration transferred to the quantum circuit. The quantum computing problem QCPROB is then solved using the configured quantum circuit QC in a solution step SOL.

[0029] In a further embodiment not specifically shown, several ideal configurations are used to solve the quantum computing problem QCPROB, with each of the ideal configurations being a configuration for solving the quantum computing problem QCPROB under the assumption of fault-free qubits. The several ideal configurations differ in the complexity of the quantum computing problem QCPROB that the ideal configurations can each solve. For each of the ideal configurations, the required error correction qubits are determined, and the total number of qubits required to realize the ideal configuration and the error correction qubits is determined, so that several configuration candidates are available for configuring the configurable quantum circuit QC.The configuration candidate that maximizes the available qubits of the quantum circuit QC is then selected, and the configurable quantum circuit QC is configured using this configuration candidate. In this embodiment, the quantum computing problem QCPROB is then solved using the quantum circuit QC configured in this way.

Claims

Patent claims 1. A method for solving a quantum computing problem, in which a configurable quantum circuit (QC) with an arrangement of qubits is used and a quantum computing problem (QCPROB) is accepted, wherein a solution algorithm is determined depending on the accepted quantum computing problem, and depending on the solution algorithm and an error susceptibility of the solution algorithm, such a configuration of qubits is determined which has a number of qubits for error correction depending on the error susceptibility of the solution algorithm, and wherein the configurable quantum circuit (QC) is configured with the determined configuration and the quantum computing problem is solved (SOL) using the configured quantum circuit (QC).

2. Method according to the preceding claim, wherein the susceptibility of the solution algorithm to errors of individual steps of the solution algorithm is determined and the configuration of the qubits is determined depending on the susceptibility of the solution algorithm to errors of the individual steps.

3. Method according to one of the preceding claims, wherein the configuration for different logical qubits for implementing the solution algorithm comprises a different number of physical qubits.

4. Method according to one of the preceding claims, wherein such a configuration is determined which exhausts an available quantity of qubits of the quantum circuit (QC).

5. A method according to any of the preceding claims, wherein a measure for the error-free nature of the solution is specified and a configuration is determined which takes into account or achieves the measure for the error-free nature of the solution. fills .

6. Method according to one of the preceding claims, wherein a measure for the complexity of the quantum computer problem is specified and the configuration is determined depending on the measure for the complexity.

7. Quantum computer with a configurable quantum circuit with an array of qubits, wherein the quantum computer has an interface for receiving a quantum computer problem (QCPROB) and wherein the quantum computer (QC) has a computing device for configuring the quantum circuit, which is configured to determine a solution algorithm depending on the received quantum computer problem (QCPROB) and, depending on the solution algorithm and an error susceptibility of the solution algorithm, to determine such a configuration of the quantum circuit that has a number of qubits for error correction depending on the error susceptibility of the solution algorithm, wherein the quantum computer (QC) is configured to configure the quantum circuit according to the determined configuration.

8. Quantum computer according to the preceding claim, configured to perform a method according to any one of claims 1 to 7.

9. Method or quantum computer in which the quantum computer problem is a manufacturing problem and / or a maintenance problem and / or a logistics problem and / or a medical image recognition problem.

Images