Quantum bit calibration method and apparatus, and quantum chip and quantum computer

Abstract

A quantum bit calibration method and apparatus, and a quantum chip and a quantum computer, which are applied in the field of quantum computing. The method comprises: acquiring a first calibration graph, the first calibration graph comprising N nodes and edges between the N nodes, wherein the edges are used for representing dependency relationships between the N nodes, each node among the N nodes is used for representing a calibration process for at least one target quantum bit, and the dependency relationships in the first calibration graph are determined on the basis of partial dependencies between target quantum bits corresponding to the N nodes, N being a positive integer (320); optimizing the dependency relationships in the first calibration graph to obtain a second calibration graph, wherein the completeness of dependency relationships in the second calibration graph is higher than the completeness of the dependency relationships in the first calibration graph (340); and on the basis of the second calibration graph, calibrating the target quantum bits (360). The method can improve the quantum bit calibration efficiency.

Description

Method and device for calibrating quantum bits, quantum chip and quantum computer

[0001] The present application claims priority to the Chinese patent application No. 202410729317.2, filed on June 5, 2024, and entitled "Method and device for calibrating quantum bits, quantum chip and quantum computer", the entire content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of quantum computing, and in particular to a method and device for calibrating quantum bits, a quantum chip and a quantum computer. BACKGROUND

[0003] A quantum computer is a physical device that performs high-speed mathematical and logical operations, storage and processing of quantum information according to the laws of quantum mechanics. A quantum bit is the basic unit of a quantum computer, and its data storage and operation control need to be based on accurate control of quantum states. The standardization of this control process is the calibration process of the quantum bit, and the calibrated quantum bit can maintain accurate control within a certain range of accuracy over a period of time.

[0004] In related technologies, a quantum calibration officer needs to specify the dependencies between each target quantum bit to be calibrated, manually determine a calibration graph, and then the quantum computer can calibrate the target quantum bits based on the calibration graph.

[0005] However, the related art method requires human intervention, resulting in low calibration efficiency of quantum bits. SUMMARY

[0006] The present application provides a method and device for calibrating quantum bits, a quantum chip and a quantum computer, which can improve the calibration efficiency of quantum bits. The technical solution is as follows:

[0007] According to an aspect of the present application, a method for calibrating quantum bits is provided, the method is executed by a quantum computer, and the method comprises:

[0008] obtaining a first calibration graph, the first calibration graph comprising N nodes and edges between the N nodes, the edges being used to represent the dependencies between the N nodes, each node in the N nodes being used to represent a calibration process for at least one target quantum bit, the dependencies in the first calibration graph being determined based on a part of the dependencies of the target quantum bits corresponding to the N nodes, N being a positive integer;

[0009] optimize the dependency relationship in the first calibration graph to obtain a second calibration graph, integrity of the dependency relationship in the second calibration graph is higher than that of the first calibration graph;

[0010] calibrate the target quantum bit based on the second calibration graph.

[0011] According to another aspect of the present application, a calibration device of a quantum bit is provided, the device comprising:

[0012] an obtaining module, configured to obtain a first calibration graph, the first calibration graph comprising N nodes and edges between the N nodes, the edges being used to represent dependency relationships between the N nodes, each of the N nodes being used to represent a calibration process on at least one target quantum bit, the dependency relationship in the first calibration graph being determined based on partial dependency of the target quantum bits corresponding to the N nodes, N being a positive integer;

[0013] an optimizing module, configured to optimize the dependency relationship in the first calibration graph to obtain a second calibration graph, integrity of the dependency relationship in the second calibration graph is higher than that of the first calibration graph;

[0014] a calibration module, configured to calibrate the target quantum bit based on the second calibration graph.

[0015] According to another aspect of the present application, a quantum chip is provided, the quantum chip comprising programmable logic circuits or program instructions, when the quantum chip is running on a quantum computer, the quantum chip is used to implement the calibration method of the quantum bit as described above based on the programmable logic circuits or program instructions.

[0016] According to another aspect of the present application, a quantum computer is provided, the quantum computer comprising a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the calibration method of the quantum bit as described above.

[0017] According to another aspect of the present application, a computer readable storage medium is provided, the computer readable storage medium storing a computer program, the computer program being loaded and executed by a processor to implement the calibration method of the quantum bit as described above.

[0018] According to another aspect of the present application, a computer program product is provided, the computer program product comprising computer instructions stored in a computer readable storage medium, a processor obtaining the computer instructions from the computer readable storage medium, so that the processor is loaded and executed to implement the calibration method of the quantum bit as described above.

[0019] The technical scheme provided by the embodiments of the present application has at least the following beneficial effects:

[0020] In the case that the existing dependency relationship in the first calibration graph is incomplete, the second calibration graph used in actual calibration is obtained by optimizing the dependency relationship in the first calibration graph, so that the dependency relationship in the second calibration graph is more complete. The entire process does not require human intervention and does not require manual specification of the dependency between each target quantum bit and determination of the second calibration graph, thereby reducing the labor cost, realizing automatic generation of the second calibration graph, and improving the generation efficiency and accuracy of the second calibration graph. Therefore, the calibration efficiency is greatly improved when the target quantum bit is calibrated based on the second calibration graph. BRIEF DESCRIPTION OF DRAWINGS

[0021] FIG. 1 is a schematic diagram of a quantum computer system provided by an example embodiment of the present application;

[0022] FIG. 2 is a schematic diagram of a quantum bit calibration method provided by an example embodiment of the present application;

[0023] FIG. 3 is a flowchart of a quantum bit calibration method provided by an example embodiment of the present application;

[0024] FIG. 4 is a flowchart of a quantum bit calibration method provided by an example embodiment of the present application;

[0025] FIG. 5 is a flowchart of a quantum bit calibration method provided by an example embodiment of the present application;

[0026] FIG. 6 is a schematic diagram of a quantum bit calibration method provided by an example embodiment of the present application;

[0027] FIG. 7 is a schematic diagram of a first calibration graph provided by an example embodiment of the present application;

[0028] FIG. 8 is a schematic diagram of a second calibration graph provided by an example embodiment of the present application;

[0029] FIG. 9 is a schematic diagram of a second calibration graph provided by an example embodiment of the present application;

[0030] FIG. 10 is a schematic diagram of a second calibration graph provided by an example embodiment of the present application;

[0031] FIG. 11 is a schematic diagram of a second calibration graph provided by an example embodiment of the present application;

[0032] FIG. 12 is a block diagram of a quantum bit calibration apparatus provided by an example embodiment of the present application;

[0033] FIG. 13 is a structural block diagram of a quantum computer provided by an example embodiment of the present application;

[0034] Figure 14 is a structural block diagram of a quantum computer according to an example embodiment of the present application. DETAILED DESCRIPTION

[0035] First, the terms involved in the embodiments of the present application are briefly introduced:

[0036] Quantum Computer: is a class of physical devices that perform high-speed mathematical and logical operations, storage and processing of quantum information in accordance with the laws of quantum mechanics. When a certain physical device processes and calculates quantum information and runs quantum algorithms, it is a quantum computer. Based on the superposition principle and quantum entanglement of quantum mechanics, quantum computers have strong parallel processing capabilities and can solve some problems that are difficult for classical computers to calculate.

[0037] Quantum Computation: is a computing method based on quantum logic. It is usually executed by a quantum computer, and the basic unit of data storage of a quantum computer is: a quantum bit (Qubit), also known as a quantum bit.

[0038] Quantum Bit (Qubit): or quantum bit, is the basic unit of data storage of a quantum computer and the basic unit of quantum computation. In a conventional computer, 0 and 1 are used as the basic units of binary, i.e., the basic units in the conventional computer are either in the "0" state or in the "1" state. However, a quantum computer can process 0 and 1 simultaneously, and the basic unit in a quantum computer can be in a linear superposition state of "0" and "1" state: |ψ>=α|0>+β|1>. Where α, β represent the complex probability amplitude of the quantum system on 0 and 1. |α| 2 , |β| 2 represent the probabilities of the quantum system being in 0 and 1, respectively.

[0039] Quantum Chip: is the central processing unit of a quantum computer. For example, a superconducting quantum chip is the central processing unit of a superconducting quantum computer, which is an electronic circuit based on superconducting Josephson junctions. The zero resistance characteristic of superconducting quantum bits and the manufacturing process close to integrated circuits make the quantum computing system constructed using superconducting quantum bits one of the most promising systems for realizing practical quantum computing.

[0040] Qubit Calibration: Qubit is a structure of storage and calculation, its data storage and operation control need to be based on the accurate control of quantum state, the standardization of this control process is the calibration process of qubit, the calibrated qubit can maintain accurate control within a certain range of accuracy for a period of time. Unlike the gate logic of digital computer, qubit is essentially an analog computing control.

[0041] Qubit Calibration Graph: It is a directed acyclic graph (DAG) used in the calibration process of qubit, which is used to graphically represent the calibration process of qubit. Graphical representation refers to the form of visual elements such as graphics and images to visually represent. Qubit needs to go through a series of calibration procedures based on physical model to meet the needs of formal calculation, usually including searching for the working frequency of quantum device, calibrating the quantum state of isolated qubit and quantum gate, and calibrating the multi-bit gate. A series of physical parameter calibration and setting links can be abstracted as a directed graph structure. The calibration process may appear repeated calibration phenomenon before and after the calibration link due to its simulation characteristics. In the currently published calibration scheme, this phenomenon is decomposed into multiple isolated nodes for different calibration targets to avoid the appearance of ring structure in the qubit calibration graph, so that it is a directed acyclic graph, and the definition of qubit calibration graph may be more accurate. Generally, a directed acyclic graph is composed of nodes and edges between nodes with single arrow, for example, node A points to node B through directed edge, then the variable corresponding to node B depends on the variable corresponding to node A, node A is the parent node of node B, and node B is the child node of node A.

[0042] First Calibration Graph: It is the initial DAG graph used in the calibration process of qubit, and the first calibration graph is also called the first qubit calibration graph. The first calibration graph needs to be expanded in more detail to obtain the second calibration graph, and the more detailed expansion includes expanding based on the dependency between the target qubits of single-bit gate and / or double-bit gate.

[0043] For example, the first calibration graph includes N nodes and edges between the N nodes, and the edges are directed edges with one-way arrows. N is a positive integer. For each node, each node corresponds to at least one target qubit, and the at least one target qubit is at least one of target qubits belonging to the same batch of processing tasks, target qubits with similar attributes, or target qubits that can perform similar calibration operations. Each node is used to represent a calibration process for the at least one target qubit, for example, for each node, it can be that the same calibration method is used to perform calibration on the target qubits in the node, or it can be that the same qubit parameter of the target qubits in the node is calibrated. At least a part of the same target qubits can be included in different nodes. For example, node A includes q4 and q5, and node B includes q5 and q7. It can be understood that no same target qubits can be included in different nodes. The edges are used to represent the dependency relationship between the nodes, and the dependency relationship is determined based on the dependency between the target qubits corresponding to the nodes. The edges are also used to represent the execution order of the calibration processes corresponding to the nodes. For example, there is an edge from node A to node B, and the calibration process of node A is performed first to calibrate the target qubits in node A. After the calibration of node A is completed, the calibration process of node B is performed to calibrate the target qubits in node B. Based on the dependency relationship between the N nodes, the N nodes include a start node and an end node, the start node is a node that has no parent node and has child nodes, and the end node is a node that has no child node and has a parent node. For example, referring to (1) shown in FIG. 2, the start node of the first calibration graph 200 is node 210, and the end node is node 220.

[0044] For at least one of the N nodes, the at least one node further comprises a plurality of sub-nodes and a plurality of sub-edges between the plurality of sub-nodes, and the sub-edges are directed sub-edges with one-way arrows. For each sub-node, the sub-node corresponds to at least one target qubit, specifically one target qubit or two target qubits. The sub-edges are used to represent the dependency relationship between the sub-nodes, and the dependency relationship is determined based on the dependency between the target qubits corresponding to the sub-nodes. The sub-edges are also used to represent the execution order of the sub-calibration processes of different target qubits in the same node. For example, in a node, there is a sub-edge from sub-node C to sub-node D, and the sub-calibration process of sub-node C is executed first to calibrate the target qubit corresponding to sub-node C. After the calibration of sub-node C is completed, the calibration process of sub-node D is executed to calibrate the target qubit corresponding to sub-node D. Based on the dependency relationship between the plurality of sub-nodes in a node, the plurality of sub-nodes include a starting sub-node and an end sub-node, the starting sub-node is a sub-node that has child sub-nodes and has no parent sub-node, and the end sub-node is a sub-node that has a parent sub-node and has no child sub-node. For example, referring to (1) shown in FIG. 2, for the node 210 in the first calibration graph 200, the node 210 includes a plurality of sub-nodes, the starting sub-nodes are sub-node 1 and sub-node 2, and the end sub-node is sub-node 5. For the node 220 in the first calibration graph 200, the node 220 includes a plurality of sub-nodes, the starting sub-nodes are sub-node 6 and sub-node 7, and the end sub-node is sub-node 9.

[0045] Calibration automation: In the calibration process of qubits, the calibration process of a plurality of groups of quantum devices with a dependency relationship can be defined as a qubit calibration graph. Each actual node in these calibration processes involves one or more qubits to be calibrated and complex control, and the control parameters have the characteristics of high dimension, multiple dependencies, and random drift. The automation of the control process of these control parameters means the automation of calibration.

[0046] FIG. 1 is a structural block diagram of a quantum computer system 100 according to an example embodiment of the present application. The quantum computer system 100 can be implemented as a system architecture of a qubit calibration method. The quantum computer system 100 includes a quantum computing terminal 120 and a quantum computer 140.

[0047] The quantum computing terminal 120 is a terminal that complies with the laws of quantum mechanics and has a quantum computing function, such as at least one of a mobile phone, a tablet computer, a vehicle terminal (car machine), a wearable device, a PC (Personal Computer), and an unmanned terminal. In some embodiments, the quantum computing terminal 120 includes a quantum chip. The quantum computing terminal 120 can install and run a client of a target application, which can be an application that provides a quantum computing function. The form of the target application is not limited in the present application, including but not limited to an App (Application) installed in the quantum computing terminal 120, a mini program, and the like, and can also be in the form of a web page.

[0048] The quantum computer 140 is a kind of physical device that complies with the laws of quantum mechanics to perform high-speed mathematical and logical operations, storage, and processing of quantum information. When a certain physical device processes and computes quantum information and runs a quantum algorithm, it is a quantum computer 140. In some embodiments, the quantum computer 140 includes a quantum chip. The quantum computer 140 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, a cloud server providing cloud computing services, a cloud database, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, and the like. In some embodiments, the quantum computer 140 can also be a background server of the target application of the quantum computing terminal 120, for providing background services for the client of the target application.

[0049] The cloud technology refers to a kind of hosting technology that unifies a series of resources such as hardware, software, and network to realize data calculation, storage, processing, and sharing in a wide area network or a local area network. The cloud technology is based on network technology, information technology, integration technology, management platform technology, application technology, and the like applied in the cloud computing business model, and can form a resource pool for on-demand use and flexible convenience. Cloud computing technology will become an important support. The background service of a technical network system requires a large amount of computing and storage resources, such as video websites, picture websites, and more portal websites. With the high development and application of the Internet industry, in the future, every item may have its own identification mark and needs to be transmitted to the background system for logical processing. Different levels of data will be processed separately, and various industry data needs strong system support, which can only be realized through cloud computing.

[0050] In some embodiments, the quantum computer 140 can also be implemented as a node in a blockchain system. The blockchain is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm. The blockchain is essentially a decentralized database, and is a series of data blocks associated using cryptographic methods, each data block containing information about a batch of network transactions, for verifying the validity of the information (anti-fake) and generating the next block. The blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.

[0051] In some embodiments, the quantum computing terminal 120 and the quantum computer 140 can also communicate through a network, such as a wired or wireless network.

[0052] The execution subject of each step of the quantum bit calibration method provided in the embodiments of the present application can be the quantum computer 140. Alternatively, the quantum computer 140 can be a standalone quantum computer, and the quantum bit calibration method is executed by the standalone quantum computer alone. The quantum computer 140 can also be a quantum computer cluster composed of multiple quantum computers, and the quantum bit calibration method is executed by the multiple quantum computers in the quantum computer cluster in interaction and cooperation, each quantum computer executing one or more steps of the quantum bit calibration method. The embodiments of the present application do not limit the number and type of the quantum computer 140. In some embodiments, the execution subject of each step of the quantum bit calibration method provided in the embodiments of the present application can also be a quantum chip in the quantum computing terminal 120 and / or the quantum computer 140.

[0053] In the related art, in the calibration automation scheme of the quantum bit, a DAG graph structure is used to drive the calibration process of the complex open quantum bit, and the DAG graph structure can support the continuous evolution of the experimental scheme, and the experimental stage is used to determine the calibration function and the corresponding quantum bit. The calibration function is used to measure and calibrate the quantum bit parameter of the quantum bit. In the calibration process of the quantum bit, the experimental scheme can be divided into two stages: the first stage: single-bit gate calibration scheme of each quantum bit; the second stage: double-bit gate calibration scheme of adjacent multi-bits. The single-bit gate calibration template of the first stage is usually defined as a DAG graph, and the object of action is a single quantum bit on a quantum chip. The double-bit gate calibration template of the second stage is defined as another DAG graph, and the object of action is a group of adjacent quantum bits on a quantum chip. Before the DAG graph construction of the first stage and the second stage, pre-configuration needs to be performed on the calibration function, for example, the target quantum bit, the related quantum bit, and the quantum bit group of the calibration function are defined, and the quantum bit group includes a plurality of target quantum bits that can be parallel. When the calibration is scheduled, the quantum calibration officer usually performs grouping on the calibration function nodes corresponding to the calibration function according to his own physical calibration experience, and further constrains the dependency relationship. That is, in the related art, the quantum calibration officer needs to specify all dependencies between the target quantum bits to be calibrated, and manually determine the final calibration graph. Next, the quantum computer calibrates the target quantum bit based on the calibration graph.

[0054] Based on this, the embodiments of the present application provide a general pre-configured first calibration graph to meet the experimental needs of the quantum calibration officer, and at the same time, provide a way to optimize the pre-configured first calibration graph to a second calibration graph used in actual calibration, increase the diversity of the calibration scheme generation of the quantum bit, meet the actual calibration needs of the quantum bit, retain the openness of the calibration automation, and do not need manual participation, and improve the calibration efficiency of the quantum bit.

[0055] FIG. 2 is a schematic diagram of a quantum bit calibration method provided by an example embodiment of the present application. Taking the method applied to the quantum computer 140 shown in FIG. 1 as an example, the steps of the quantum bit calibration method executed by the quantum computer 140 are briefly described as follows:

[0056] Step 1. The quantum computer 140 acquires a first calibration graph 200, the first calibration graph 200 includes N nodes and edges between the N nodes, the edges are used to represent the dependency relationship between the N nodes, and each node in the N nodes is used to represent a calibration process of at least one target quantum bit. The dependency relationship in the first calibration graph 200 is determined based on a part of the dependency of the target quantum bits corresponding to the N nodes, and N is a positive integer.

[0057] Optionally, the target qubits are qubits to be calibrated. In an example, the target qubits are represented by letters q and numbers. The first calibration graph 200 is obtained in combination with the current dependencies between the target qubits of the quantum computer, and based on a general standard template, which is pre-configured and determined to basically meet the experimental needs of the quantum calibration specialist. The standard template is used to indicate the calibration process of a plurality of target qubits and the execution order between the calibration processes. The execution order includes at least one of sequential execution and parallel execution, which can be set according to actual technical needs. Specifically, the standard template defines batch nodes and edges between the batch nodes, wherein for each batch node, each batch node corresponds to a plurality of target qubits, each batch node is used to represent a calibration process for the plurality of target qubits, and the plurality of target qubits can be at least one of target qubits belonging to the same batch task, target qubits having similar properties, or target qubits that can perform similar calibration operations. The batch nodes in the above-mentioned standard template correspond to the nodes in the first calibration graph 200, and the edges between the batch nodes correspond to the edges in the first calibration graph 200. Therefore, the dependencies in the first calibration graph 200 obtained based on the standard template are not complete, and are only used to represent the dependencies of a part of the target qubits.

[0058] Referring to (1) shown in FIG. 2, assuming N = 2, the first calibration graph 200 includes 2 nodes: node 210 and node 220, and an edge between node 210 and node 220 is used to represent the dependency relationship between node 210 and node 220, the edge from node 210 to node 220 indicates that node 220 depends on node 210, so the calibration process represented by node 210 is performed first to calibrate the target qubits in node 210, and after the calibration of the target qubits in node 210 is completed, the calibration process represented by node 220 is performed to calibrate the target qubits in node 220. The target qubits in node 210 include: q1, q2, q3, q4, q5; the target qubits in node 220 include: q4, q5, q7, q8, q9. The node also includes a plurality of sub-nodes and a plurality of sub-edges between the sub-nodes, and the sub-edges represent the dependency relationship between the sub-nodes. The 5 sub-nodes in node 210 are: sub-node 1, sub-node 2, sub-node 3, sub-node 4, sub-node 5, and the sub-edges between the 5 sub-nodes are: sub-edge 13 between sub-node 1 and sub-node 3, sub-edge 24 between sub-node 2 and sub-node 4, sub-edge 35 between sub-node 3 and sub-node 5, and sub-edge 45 between sub-node 4 and sub-node 5. Node 220 also includes 4 sub-nodes: sub-node 6, sub-node 7, sub-node 8, and sub-node 9, node 6 corresponds to q4 and q5, node 7 corresponds to q7, node 8 corresponds to q8, and node 9 corresponds to q9, and the sub-edges between the 4 sub-nodes are: sub-edge 68 between sub-node 6 and sub-node 8, sub-edge 78 between sub-node 7 and sub-node 8, and sub-edge 89 between sub-node 8 and sub-node 9. The dependency relationship in the first calibration graph 200 shown in (1) of FIG. 2 is incomplete and does not represent the dependency relationship between the sub-nodes in node 210 and the sub-nodes in node 220. Therefore, when the quantum computer 140 calibrates the target qubits based on the first calibration graph 200, the calibration process can only take node 210 as the starting node and perform the calibration of the target qubits based on the dependency relationship in the first calibration graph 200.

[0059] Step 2. The quantum computer 140 optimizes the dependency relationship in the first calibration graph 200 to obtain a second calibration graph 300, and the completeness of the dependency relationship in the second calibration graph 300 is higher than that of the first calibration graph 200.

[0060] Optionally, as shown in (2) of FIG. 2, the dependency relationship in the first calibration graph 200 is optimized by adding a new edge corresponding to a newly added dependency relationship in the first calibration graph 200. A sub-edge 56 can be added between the sub-node 5 and the sub-node 6, so that the dependency relationship in the second calibration graph 300 is more complete. It should be noted that (2) of FIG. 2 only shows one way to optimize the dependency relationship in the first calibration graph 200. According to actual technical needs, one or more other sub-edges can also be added to the first calibration graph 200, which is not limited in the present embodiment. Based on this, when the quantum computer 140 calibrates the target qubits based on the second calibration graph 300, the calibration process can be to take the sub-node 1 and / or the sub-node 2 and / or the sub-node 7 as a parallel starting sub-node, and perform calibration of the target qubits based on the dependency relationship in the second calibration graph 300.

[0061] Step 3. The quantum computer 140 calibrates the target qubits based on the second calibration graph 300.

[0062] Optionally, the quantum computer 140 sequentially performs calibration operations on the target qubits based on the dependency relationship in the second calibration graph 300, and calibrates the quantum bit parameters of the target qubits through the calibration operations to calibrate the target qubits.

[0063] In summary, the quantum bit calibration method provided by the embodiments of the present application, the quantum computer obtains a first calibration graph, the first calibration graph is determined in combination with the dependency between the current target qubits of the quantum computer and based on a general standard template, the dependency relationship in the first calibration graph is determined based on part of the dependency of the target qubits, the existing dependency relationship in the first calibration graph is incomplete, the dependency relationship in the first calibration graph is optimized, and a second calibration graph used in actual calibration is obtained by conversion, the second calibration graph is a more fine-grained expansion of the first calibration graph, the dependency relationship in the second calibration graph is more complete, the above entire process does not require human intervention and does not require human to specify the dependency between each target qubit and determine the second calibration graph, which reduces the labor cost and realizes automatic generation of the second calibration graph, and further improves the generation efficiency and accuracy of the second calibration graph. Therefore, when the target qubits are calibrated based on the second calibration graph, the calibration efficiency is greatly improved.

[0064] FIG. 3 is a flowchart of a quantum bit calibration method provided by an example embodiment of the present application. This method is taken as an example for application to the quantum computer 140 shown in FIG. 1. The method includes at least part of the steps in steps 320, 340 and 360:

[0065] At step 320, a first calibration graph is obtained, the first calibration graph comprising N nodes and edges between the N nodes, the edges being used to represent dependencies between the N nodes, each of the N nodes being used to represent a calibration process on at least one target qubit, the dependencies in the first calibration graph being determined based on a portion of dependencies of the target qubits corresponding to the N nodes, N being a positive integer.

[0066] The first calibration graph is an initial DAG graph used in the calibration process of the qubits, and the first calibration graph is also referred to as a first qubit calibration graph. The first calibration graph needs to be expanded to a second calibration graph in a more fine-grained manner, and the more fine-grained expansion includes expansion based on dependencies between target qubits of single-bit gates and / or double-bit gates. This will be described in detail in subsequent embodiments.

[0067] In some embodiments, the first calibration graph is determined in combination with current dependencies between target qubits of the quantum computer and based on a general standard template, and the general standard template is pre-configured and determined to basically meet the experimental needs of the quantum calibration specialist. Specifically, the standard template defines batch nodes and edges between the batch nodes, wherein each batch node is used to represent a plurality of target qubits that can be parallel in the calibration process, and the plurality of target qubits in each batch node can be at least one of target qubits belonging to the same batch task, target qubits having similar properties, or target qubits that can perform similar calibration operations. The batch nodes in the standard template correspond to the nodes in the first calibration graph, and the edges between the batch nodes correspond to the edges in the first calibration graph, and therefore the dependencies in the first calibration graph obtained based on the standard template are not complete. In the subsequent process, the first calibration graph also needs to be further optimized to meet the actual calibration needs of the qubits.

[0068] The first calibration graph comprises N nodes and edges between the N nodes, and the edges are used to represent dependencies between the N nodes. N is a positive integer. Exemplarily, the edges are directed edges, one edge is used to connect two different nodes, and the direction of one edge is used to represent the dependency between the two nodes. For example, the calibration result or data of node A needs to be transmitted to node B, and in the first calibration graph, there is an edge from node A to node B, and node B depends on node A.

[0069] The target qubits are qubits to be calibrated. The target qubits include one or more.

[0070] In some embodiments, the target qubit is determined experimentally. Illustratively, a quantum computer or a quantum chip includes a plurality of qubits, each qubit having a set of qubit parameters, which can be at least one of: a readout pulse frequency, a readout pulse length, a readout pulse power, a qubit frequency, a qubit relaxation time, a qubit decoherence time, a fidelity of a single gate operation. Before the quantum computer or the quantum chip is put into use, a series of experiments need to be performed to obtain the set of qubit parameters for each qubit. When there is at least one qubit parameter of a qubit that drifts, i.e., at least one qubit parameter is out of the corresponding acceptable range, the qubit needs to be calibrated, and the qubit is the target qubit at this time. Alternatively, a neural network model is used to determine whether a qubit is a target qubit based on the qubit parameters of the qubit. The neural network model is trained using sample data, which includes sample qubits and their corresponding sample qubit parameters, and the sample qubits include label information of sample target qubits.

[0071] The target qubits have dependencies. Optionally, the dependencies between the target qubits can be predefined, or can also be characterized by at least one of the state, coherence, and degree of mutual influence of the target qubits. Specifically, the state of the target qubit includes: being in a "0" state or a "1" state, being in a linear superposition of "0" and "1" states.

[0072] In other embodiments, the dependencies between the target qubits can also be characterized by at least one of the dependencies between the relevant target qubits and the qubit groups. The relevant qubits include at least one of the target qubits that repeat, the target qubits that are physically less than a threshold distance apart, and the target qubits that influence each other. A qubit group includes a plurality of target qubits that can be parallel. In one example, the plurality of target qubits that can be parallel can be at least one of the target qubits that belong to the same batch of processing tasks, the target qubits that have similar properties, or the target qubits that can perform similar calibration operations.

[0073] In some embodiments, there can be dependencies between the quantum bit parameters of the one or more target quantum bits. For example, one quantum bit parameter of a target quantum bit depends on at least one other quantum bit parameter and the calibration of the at least one other quantum bit parameter. For example, the quantum bit frequency of a target quantum bit needs to be calibrated before the target quantum bit is calibrated using a certain drive. Or, a series of quantum bit parameters of a target quantum bit needs to be calibrated before a series of quantum bit parameters of the next target quantum bit is calibrated. The dependencies between the target quantum bits can also be characterized by the dependencies existing between the quantum bit parameters of the target quantum bits.

[0074] Optionally, in the first calibration graph, each node of the N nodes corresponds to at least one target quantum bit, N being a positive integer. Each node is then used to characterize one calibration process of the at least one target quantum bit. For example, for each node, it can be that the same calibration method is used to perform calibration on the target quantum bits in the node, or it can be that the same quantum bit parameter of the target quantum bits in the node is calibrated. The number of target quantum bits corresponding to different nodes of the N nodes can be the same or different. For example, assuming N = 2, the first calibration graph includes 2 nodes: node 1 and node 2, then node 1 can correspond to 5 target quantum bits, and node 2 can also correspond to 5 target quantum bits, or node 2 can also correspond to 4 target quantum bits. At least some of the target quantum bits can be included in different nodes. It can be understood that no target quantum bits can be included in different nodes.

[0075] The dependencies are determined based on the dependencies between the target quantum bits. That is, the dependencies are also used to characterize the dependencies between the target quantum bits. Optionally, the dependencies include inter-node dependencies, intra-node dependencies, and cross-node dependencies. The inter-node dependencies are determined based on the dependencies between the target quantum bits corresponding to different nodes. The intra-node dependencies are determined based on the dependencies between the target quantum bits corresponding to different sub-nodes within the same node. The cross-node dependencies are determined based on the dependencies between the target quantum bits corresponding to two sub-nodes belonging to different nodes. For the first calibration graph, since the first calibration graph is determined in combination with the current dependencies between the target quantum bits of the quantum computer and based on the standard template, the first calibration graph contains the inter-node dependencies and the intra-node dependencies, but does not contain the cross-node dependencies, that is, the dependencies in the first calibration graph are determined based on a part of the dependencies between the target quantum bits corresponding to the N nodes, that is, the dependencies in the first calibration graph are incomplete, or it can be understood that the completeness of the dependencies in the first calibration graph is lower than a threshold.

[0076] In some embodiments, the dependencies in the first calibration graph include at least one of inter-node dependencies and intra-node dependencies. An inter-node dependency is a dependency between different nodes in the N nodes. For example, referring to (1) in FIG. 2, the inter-node dependencies are the dependencies between node 210 and node 220. There are at least one node in the N nodes, which further includes a plurality of sub-nodes and sub-edges between the plurality of sub-nodes. An intra-node dependency is a dependency between different sub-nodes in the same node in the N nodes. For example, referring to (1) in FIG. 2, the intra-node dependencies are the dependencies between different sub-nodes in node 210 and the dependencies between different sub-nodes in node 220. Then the dependencies in the first calibration graph are incomplete and do not represent cross-node dependencies, which are dependencies between two sub-nodes belonging to different nodes in the N nodes. For example, referring to (1) in FIG. 2, the cross-node dependencies include the dependency between sub-node 5 in node 210 and sub-node 6 in node 220.

[0077] In some examples, the cross-node dependencies are predefined, or are specified by a quantum calibration officer, or are determined by a neural network model, or are determined by the quantum computer based on the intra-node dependencies, the inter-node dependencies, and the target qubits corresponding to the N nodes in the first calibration graph. This will be described in detail in subsequent embodiments.

[0078] At step 340, the dependencies in the first calibration graph are optimized to obtain a second calibration graph, and the completeness of the dependencies in the second calibration graph is higher than that of the first calibration graph.

[0079] The optimization is to add new dependencies for the target qubits in the first calibration graph to perfect the dependencies in the first calibration graph. Or, the optimization is to perform a more fine-grained expansion on the first calibration graph, and the more fine-grained expansion includes expansion based on the dependencies between the target qubits of the single-qubit gates and / or the two-qubit gates. The second calibration graph is the calibration graph obtained after optimizing the first calibration graph, and generally, the second calibration graph retains the existing dependencies in the first calibration graph and adds some other dependencies.

[0080] The dependencies in the first calibration graph include at least one of intra-node dependencies and inter-node dependencies. Then the cross-node dependencies are added to the first calibration graph to obtain a second calibration graph. The completeness of the dependencies in the second calibration graph is higher than that of the first calibration graph. For example, the dependencies in the second calibration graph are determined based on the full dependencies of the target qubits corresponding to the N nodes.

[0081] In some embodiments, there are different ways to optimize the dependencies in the first calibration graph. In this embodiment, the way to optimize the dependencies in the first calibration graph is referred to as an unfolding strategy, and different second calibration graphs are obtained by using different unfolding strategies. For example, a set of automatic unfolding strategies is pre-configured, and different unfolding strategies in the set of automatic unfolding strategies are used to indicate different ways to optimize the dependencies in the first calibration graph. In combination with the foregoing embodiment, different unfolding strategies correspond to different cross-node dependencies. Before step 340, the method can further include: determining a target unfolding strategy from the set of automatic unfolding strategies; and then, optimizing the dependencies in the first calibration graph based on the target unfolding strategy to obtain a second calibration graph under the target unfolding strategy.

[0082] At step 360, the target qubits are calibrated based on the second calibration graph.

[0083] Optionally, the dependencies in the second calibration graph are used to indicate the order of calibrating the target qubits, and the calibration operations are sequentially performed on the target qubits based on the dependencies in the second calibration graph to calibrate the quantum bit parameters of the target qubits by the calibration operations, so as to calibrate the target qubits. Until the output signals of the target qubits are in a normal state, the calibration of the target qubits is completed. Optionally, the calibration operation can be at least one of adjusting the correlation between the target qubits, adjusting the quantum bit parameters of the target qubits to a corresponding acceptable range, and the like.

[0084] In some embodiments, the calibration function is used to indicate the calibration manner of the target qubits, or to indicate the specific calibration operation performed on the target qubits to calibrate the target qubits. Optionally, the calibration function and the target qubits are in a many-to-many relationship, one calibration function can correspond to one or more target qubits, and one target qubit can also correspond to one or more calibration functions. That is, one calibration function can calibrate one or more target qubits, and one target qubit can also be calibrated by one or more calibration functions. The correspondence between the calibration function and the target qubit is determined based on experiments, and the experimental stage is not described herein.

[0085] Optionally, in the first calibration graph and the second calibration graph, the calibration functions correspond to the nodes and the sub-nodes in the nodes in multiple ways. Correspondence mode 1: one node corresponds to one calibration function, and multiple sub-nodes in one node correspond to the same calibration function. At this time, the node is also called a calibration function node or a calibration node. In correspondence mode 1, when the first calibration graph is unfolded to obtain the second calibration graph, the calibration function corresponding to each sub-node in the second calibration graph can also be further adjusted and modified. Correspondence mode 2: one sub-node corresponds to one calibration function, and one node corresponds to a type of calibration function. At this time, the sub-node is also called a calibration function node. Regardless of which correspondence mode, the target qubits in the sub-nodes all have corresponding calibration functions. Then, based on the dependency relationship in the second calibration graph and according to the calibration function corresponding to the target qubit, the calibration operation is sequentially performed on the target qubit, the quantum bit parameter of the target qubit is calibrated through the calibration operation, and the target qubit is calibrated. Until the output signal of the target qubit is in a normal state, the calibration of the target qubit is completed.

[0086] In summary, the calibration method of the quantum bit provided by the embodiments of the present application, the quantum computer obtains a first calibration graph, the first calibration graph includes N nodes and edges between the N nodes, the edges are used to represent the dependency relationship between the N nodes, each node in the N nodes is used to represent a calibration process for at least one target qubit, the dependency relationship in the first calibration graph is determined based on part of the dependency of the target qubits corresponding to the N nodes, and N is a positive integer; the dependency relationship in the first calibration graph is optimized to obtain a second calibration graph, and the completeness of the dependency relationship in the second calibration graph is higher than that of the first calibration graph; and the target qubit is calibrated based on the second calibration graph. Accordingly, since the existing dependency relationship in the first calibration graph is incomplete, the second calibration graph used in actual calibration is obtained by optimizing the dependency relationship in the first calibration graph, so that the dependency relationship in the second calibration graph is more complete, and the second calibration graph can meet the actual calibration requirements of the quantum bit. The above entire process does not require human participation and does not require human to specify the dependency between each target qubit and determine the second calibration graph, thereby reducing the labor cost, realizing the automatic generation of the second calibration graph, and improving the generation efficiency and accuracy of the second calibration graph. Therefore, when the target qubit is calibrated based on the second calibration graph, the calibration efficiency is greatly improved.

[0087] In some embodiments, the dependencies in the first calibration graph are optimized by edges between nodes, and sub-edges between sub-nodes in each node, and the optimized dependencies in the first calibration graph can be adding at least one new edge in the first calibration graph, so that the second calibration graph is a more fine-grained expansion of the first calibration graph. FIG. 4 is a flowchart of a calibration method of a quantum bit according to an example embodiment of the present application. Step 340 is implemented as step 400.

[0088] Step 400: adding at least one new edge in the first calibration graph to obtain a second calibration graph; wherein the at least one new edge corresponds to at least one cross-node dependency, and the at least one cross-node dependency includes a dependency between two sub-nodes belonging to different nodes in the N nodes.

[0089] The new edge refers to a newly added sub-edge in the first calibration graph. The new edge includes at least one.

[0090] In some embodiments, the dependencies in the first calibration graph include intra-node dependencies and inter-node dependencies. The at least one new edge is a sub-edge corresponding to at least one cross-node dependency. For example, the at least one new edge corresponds to at least one cross-node dependency, and the at least one cross-node dependency includes a dependency between two sub-nodes belonging to different nodes in the N nodes. In other embodiments, the new edge can also be a sub-edge between two preset sub-nodes. The preset sub-nodes can be sub-nodes in a preset position in the first calibration graph, and the two sub-nodes are directly connected by the new edge. The preset sub-nodes, preset positions, etc. can be set according to actual technical needs, and are not limited herein.

[0091] For example, referring to (1) shown in FIG. 2, assuming N = 2, the inter-node dependency is the dependency between node 210 and node 220, the intra-node dependency is the dependency between different sub-nodes in node 210 and the dependency between different sub-nodes in node 220, referring to (2) shown in FIG. 2, the cross-node dependency includes the dependency between sub-node 5 in node 210 and sub-node 6 in node 220. It should be noted that (2) in FIG. 2 only shows an example of cross-node dependency, and the cross-node dependency can also be the dependency between sub-node 4 in node 210 and sub-node 6 in node 220, or the dependency between sub-node 5 in node 210 and sub-node 7 in node 220, or a combination of the foregoing cross-node dependencies, which is not limited in the present embodiment.

[0092] In this embodiment, by adding at least one new edge in the first calibration graph to obtain the second calibration graph, the existing dependency relationship in the first calibration graph can be improved, so that the second calibration graph not only retains the existing dependency relationship in the first calibration graph, but also makes the second calibration graph more capable of meeting the actual calibration requirements of the quantum bits. Since the at least one new edge is one-to-one corresponding to the at least one cross-node dependency relationship, different new edges can be added in the first calibration graph, which also increases the diversity of the calibration scheme generation of the quantum bits, and different second calibration graphs are obtained. Therefore, when calibrating the target quantum bits based on different second calibration graphs, the calibration flexibility can also be improved.

[0093] In some embodiments, the at least one cross-node dependency relationship is predefined. For example, the cross-node dependency relationship is defined as the dependency relationship between the starting sub-nodes in different nodes in the first calibration graph. Then, when there is no dependency relationship between the starting sub-nodes in different nodes in the first calibration graph, a new edge corresponding to this cross-node dependency relationship is added in the first calibration graph. Alternatively, the at least one cross-node dependency relationship is specified by a quantum calibration officer based on experimental experience. For example, the quantum calibration officer inputs the cross-node dependency relationship in the first calibration graph in the quantum computer based on each target quantum bit contained in the first calibration graph. Alternatively, the at least one cross-node dependency relationship is determined by a neural network model. For example, the quantum bit parameters corresponding to each target quantum bit contained in the first calibration graph are input into the neural network model, the target quantum bits with dependency are determined based on the neural network model, and the at least one cross-node dependency relationship is determined based on these target quantum bits with dependency. The neural network model is trained using sample data to obtain the neural network model, and the sample data includes sample quantum bits and sample quantum bit parameters corresponding to the sample quantum bits. The sample quantum bits include label information of the dependency between the sample target quantum bits.

[0094] In some embodiments, the at least one cross-node dependency relationship is determined by the quantum computer based on the dependency relationship in the first calibration graph. Please continue to refer to FIG. 4, and step 400 is specifically implemented as step 420:

[0095] Step 420, based on at least one of the inter-node dependency relationship, the intra-node dependency relationship, and the target quantum bits corresponding to the sub-nodes in different nodes in the N nodes in the first calibration graph, at least one new edge is added in the first calibration graph to obtain the second calibration graph; wherein the inter-node dependency relationship is the dependency relationship between different nodes in the N nodes, and the intra-node dependency relationship is the dependency relationship between different sub-nodes in the same node in the N nodes.

[0096] The first calibration graph includes inter-node dependency relationships and intra-node dependency relationships. The inter-node dependency relationships are dependency relationships between different nodes of the N nodes, and the intra-node dependency relationships are dependency relationships between different sub-nodes in the same node of the N nodes.

[0097] For example, based on at least one of the inter-node dependency relationships, the intra-node dependency relationships, and the target qubits corresponding to the sub-nodes in different nodes of the N nodes in the first calibration graph, at least one cross-node dependency relationship is determined, at least one newly added edge corresponding to the at least one cross-node dependency relationship is added in the first calibration graph, and a second calibration graph is obtained.

[0098] In this embodiment, by adding at least one newly added edge in the first calibration graph based on at least one of the inter-node dependency relationships, the intra-node dependency relationships, and the target qubits corresponding to the sub-nodes in different nodes of the N nodes, a second calibration graph is obtained, which can still satisfy the existing dependency relationships in the first calibration graph, and the accuracy of the second calibration graph is improved.

[0099] In some embodiments, at least one cross-node dependency relationship is determined based on at least one expansion strategy in the automatic expansion strategy set, and the at least one expansion strategy corresponds to the at least one cross-node dependency relationship in a one-to-one manner. The following embodiments take four expansion strategies as examples to illustrate different cross-node dependency relationships determined based on different expansion strategies.

[0100] · Expansion strategy one:

[0101] The at least one cross-node dependency relationship includes a first dependency relationship, the first dependency relationship is a dependency relationship between each sub-node in a first node and each sub-node in a second node, and the first node and the second node are different nodes of the N nodes. Each sub-node in the first node can refer to all or part or each sub-node in the first node. Each sub-node in the second node can refer to all or part or each sub-node in the second node. It should be noted that the first node and the second node can be any two different nodes of the N nodes, or can be different nodes of the N nodes that have inter-node dependency relationships. For example, assuming that N=3, the three nodes in the first calibration graph are node 1, node 2, and node 3, one edge points from node 1 to node 2, and one edge points from node 2 to node 3. The first node and the second node can be any node 1 and node 3, or the first node and the second node can be node 1 and node 2 that have inter-node dependency relationships. Please continue to refer to FIG. 4, and the method further includes step 421:

[0102] Step 421, determine the first dependency relationship between each sub-node in the first node and each sub-node in the second node based on the dependency between the target qubits corresponding to each sub-node in the first node and the target qubits corresponding to each sub-node in the second node.

[0103] Each sub-node can correspond to one or more target qubits. In some embodiments, for one sub-node in the first node and one sub-node in the second node, or for two sub-nodes in the same node, when there is at least one same target qubit in the target qubits corresponding to the two sub-nodes, there is a dependency between the target qubits corresponding to the two sub-nodes. Or, a target qubit group with dependency is preset, and when there is a target qubit in the target qubits corresponding to the two sub-nodes that is in the same target qubit group, there is a dependency between the target qubits corresponding to the two sub-nodes.

[0104] For example, based on the dependency between the target qubits corresponding to each sub-node in the first node and the target qubits corresponding to each sub-node in the second node, the first dependency relationship between each sub-node in the first node and each sub-node in the second node is determined. Next, at least one new edge is added in the first calibration graph according to the first dependency relationship, that is, the two sub-nodes with the first dependency relationship are connected by a sub-edge to obtain a second calibration graph.

[0105] The manner of the present embodiment is also called target qubit dependency expansion (QcompsNameAnyMatch), and the dependency between all sub-nodes is added. Through the manner of the present embodiment, the second calibration graph is obtained by performing more fine-grained expansion on the first calibration graph from the dimension of the dependency between the target qubits corresponding to the sub-nodes, so that the second calibration graph can better reflect the dependency between the target qubits compared to the first calibration graph, and can ensure that each sub-node in the second calibration graph has a corresponding dependency relationship, avoiding missing any dependency relationship, and when performing calibration based on the second calibration graph, multiple target qubits can be calibrated in parallel in multiple optional manners, improving the calibration efficiency.

[0106] · Expansion strategy two:

[0107] The at least one cross-node dependency includes a second dependency, the second dependency being a subset of the first dependency, the second dependency being a dependency between a terminal sub-node in the first node and each sub-node in the second node, the first node and the second node being different nodes of the N nodes that have a cross-node dependency. Each sub-node in the second node can refer to all or part or each sub-node in the second node. For example, referring to (1) shown in FIG. 2, the terminal sub-node in the node 210 is sub-node 5, and the terminal sub-node in the node 220 is sub-node 9. Please continue to refer to FIG. 4, the method further includes steps 422 and 423:

[0108] In step 422, a terminal sub-node in the first node is determined from each sub-node in the first node based on the intra-node dependency of the first node.

[0109] Each sub-node in the first sub-node includes a terminal sub-node. The terminal sub-node is at least one.

[0110] Optionally, the terminal sub-node in the first node is a sub-node located at the end of the intra-node dependency of the first node. In the first node, when a sub-node is not a terminal sub-node, the sub-node has a child edge pointing to the next sub-node, then the terminal sub-node in the first node can also be understood as a sub-node in the first node that does not have a child edge pointing to the next sub-node. In other examples, for example, node A points to node B through a directed edge, then the variable B corresponding to node B depends on the variable A corresponding to node A, node A is the parent node of node B, and node B is the child node of node A. Then the terminal sub-node in the first node can also be understood as a sub-node that has a parent sub-node and does not have a child sub-node.

[0111] In step 423, the second dependency between the terminal sub-node in the first node and each sub-node in the second node is determined based on the dependency between the target qubit corresponding to the terminal sub-node in the first node and the target qubit corresponding to each sub-node in the second node.

[0112] Each sub-node can correspond to one or more target qubits. In some embodiments, for a sub-node in the first node and a sub-node in the second node, or for two sub-nodes in the same node, when there is at least one same target qubit in the target qubits corresponding to the two sub-nodes, there is a dependency between the target qubits corresponding to the two sub-nodes. Or, a target qubit group with dependency is preset, and when there is a target qubit in the same target qubit group in the target qubits corresponding to the two sub-nodes, there is a dependency between the target qubits corresponding to the two sub-nodes.

[0113] For example, the second dependency relationship between the terminal sub-node in the first node and each sub-node in the second node is determined based on the dependency between the target qubit corresponding to the terminal sub-node in the first node and the target qubit corresponding to each sub-node in the second node. When there is more than one terminal sub-node in the first node, the step is repeated to determine the second dependency relationship between each terminal sub-node in the first node and each sub-node in the second node. In the embodiment, the terminal sub-node in the first node is at least one of each sub-node in the first node, and the second dependency relationship is a subset of the first dependency relationship. Next, at least one new edge is added in the first calibration graph according to the second dependency relationship, that is, the two sub-nodes with the second dependency relationship are connected by a sub-edge to obtain a second calibration graph.

[0114] The manner of the embodiment is also referred to as target qubit dependency expansion in accordance with node pre-grouping (QcompsNameAnyMatchRespectBatch), which is used to add dependency related to the terminal sub-node in each node based on the inter-node dependency relationship between nodes and nodes. Through the manner of the embodiment, the second calibration graph is obtained by performing more fine-grained expansion on the first calibration graph from the dependency between the target qubit corresponding to the terminal sub-node in the first node and the target qubit corresponding to each sub-node in the second node, so that the second calibration graph can better reflect the dependency between the target qubit corresponding to the terminal sub-node in the first node and the target qubit corresponding to each sub-node in the second node compared to the first calibration graph, and can ensure that each node in the second calibration graph has a corresponding dependency relationship. When calibration is performed based on the second calibration graph, multiple target qubits can be calibrated in parallel in multiple optional manners, and the calibration efficiency is improved. Since the embodiment does not need to determine the corresponding dependency relationship of each sub-node, but only needs to determine the corresponding dependency relationship of the terminal sub-node, the optimization efficiency of the first calibration graph is also improved.

[0115] · Expansion strategy three:

[0116] The at least one cross-node dependency relationship includes a third dependency relationship, and the third dependency relationship is a dependency relationship between the terminal sub-node in the first node and the starting sub-node in the second node. The first node and the second node are different nodes in the N nodes that have an inter-node dependency relationship. For example, referring to (1) shown in FIG. 2, the terminal sub-node in the node 210 is the sub-node 5, and the starting sub-node in the node 220 is the sub-node 6 and the sub-node 7. Please continue to refer to FIG. 4, and the method further includes steps 424 and 425:

[0117] Step 424, determining an end sub-node in the first node from each sub-node in the first node based on the intra-node dependency of the first node, and determining a start sub-node in the second node from each sub-node in the second node based on the intra-node dependency of the second node.

[0118] Each sub-node in the first node can refer to all or part or each sub-node in the first node. Each sub-node in the first node includes the end sub-node. The end sub-node is at least one. Each sub-node in the second node can refer to all or part or each sub-node in the second node. Each sub-node in the second node includes the start sub-node. The start sub-node is at least one.

[0119] Optionally, the end sub-node in the first node is a sub-node at the end of the intra-node dependency of the first node. In the first node, when a sub-node is not the end sub-node, another sub-node has a child edge pointing to the sub-node, the end sub-node in the first node can also be understood as a sub-node in the first node that does not have a child edge pointing to another sub-node. In other examples, for example, node A points to node B through a directed edge, then the variable B corresponding to node B depends on the variable A corresponding to node A, node A is the parent node of node B, and node B is the child node of node A. The end sub-node in the first node can also be understood as a sub-node that has a parent sub-node and does not have a child sub-node.

[0120] Optionally, the start sub-node in the second node is a sub-node at the start of the intra-node dependency of the second node. In the second node, when a sub-node is not the start sub-node, another sub-node has a child edge pointing to the sub-node, the start sub-node in the second node can also be understood as a sub-node in the second node that does not have a child edge pointing to another sub-node. In other examples, for example, node A points to node B through a directed edge, then the variable B corresponding to node B depends on the variable A corresponding to node A, node A is the parent node of node B, and node B is the child node of node A. The start sub-node in the second node can also be understood as a sub-node that has a child sub-node and does not have a parent sub-node.

[0121] For example, determining an end sub-node in the first node from each sub-node in the first node based on the intra-node dependency of the first node, and determining a start sub-node in the second node from each sub-node in the second node based on the intra-node dependency of the second node.

[0122] Step 425, determine a third dependency relationship between the end sub-node in the first node and the start sub-node in the second node.

[0123] For example, without determining whether the target qubits corresponding to the end sub-node in the first node and the start sub-node in the second node respectively exist dependency, directly determine that the third dependency relationship exists between the end sub-node in the first node and the start sub-node in the second node. When there is more than one end sub-node in the first node and more than one start sub-node in the second node, repeat the step to determine the third dependency relationship between each end sub-node in the first node and each start sub-node in the second node. Next, add at least one new edge in the first calibration graph according to the third dependency relationship, that is, connect the two sub-nodes with the third dependency relationship using a sub-edge, to obtain a second calibration graph.

[0124] The manner of the embodiment is also called node pre-grouping dependency expansion (PreserveBatchEdge), which is used to add the dependency between the end sub-node in each node and the start sub-node in other nodes based on the inter-node dependency relationship between the nodes. Through the manner of the embodiment, the second calibration graph is obtained by performing more fine-grained expansion on the first calibration graph from the dimension of the end sub-node in the first node and the start sub-node in the second node, so that compared with the first calibration graph, the second calibration graph can determine the dependency relationship between the start sub-node and the end sub-node of the two nodes that exist the inter-node dependency relationship. When performing calibration based on the second calibration graph, multiple target qubits can be calibrated in parallel in multiple optional manners, improving the calibration efficiency. Since the embodiment does not need to determine the dependency between the target qubits, but only needs to determine the dependency between the end sub-node and the start sub-node, the data processing efficiency and the optimization efficiency of the first calibration graph can also be improved.

[0125] • Expansion strategy four:

[0126] In some embodiments, various expansion strategies can be used in combination according to actual technical needs to determine multiple cross-node dependency relationships and add multiple new edges corresponding to the cross-node dependency relationships in the first calibration graph to obtain a second calibration graph. Alternatively, at least one cross-node dependency relationship can be a combination of the first dependency relationship and the second dependency relationship, a combination of the first dependency relationship and the third dependency relationship, a combination of the second dependency relationship and the third dependency relationship, and a combination of the first dependency relationship, the second dependency relationship, and the third dependency relationship, which will not be described here. The embodiment can preserve the openness and freedom of generating the second calibration graph as much as possible, making the determination manner of the dependency relationship more flexible.

[0127] • First calibration graph:

[0128] In some embodiments, the first calibration graph is also referred to as a preconfigured DAG graph (CalBatchSessionGraph). The first calibration graph is determined in combination with the dependencies between the target qubits of the quantum computer at present, and based on a general standard template which is preconfigured and can basically meet the experimental needs of the quantum calibration officer. In this embodiment, the preset types of information required to be contained in the first calibration graph are pre-defined, and then the quantum computer can determine the specific information corresponding to the preset types according to the actual technical needs, to obtain the first calibration graph. FIG. 5 is a flowchart of a quantum bit calibration method provided in an exemplary embodiment of the present application. Then, step 320 can be specifically implemented as steps 520, 540 and 560:

[0129] In step 520, the dependency relationship between the calibration functions is determined based on the calibration functions corresponding to the target qubits and the dependencies between the target qubits.

[0130] The calibration function is used to indicate the calibration manner of the target qubit, or to indicate the specific calibration operation performed on the target qubit to calibrate the target qubit. Optionally, the calibration function and the target qubit are in a many-to-many relationship, one calibration function can correspond to one or more target qubits, and one target qubit can also correspond to one or more calibration functions. That is, one calibration function can calibrate one or more target qubits, and one target qubit can also be calibrated by one or more calibration functions. The correspondence between the calibration function and the target qubit is determined based on experiments, which will not be described in detail in this embodiment.

[0131] The foregoing embodiments describe that in the first calibration graph and the second calibration graph, the calibration function and the node and the sub-node in the node have multiple corresponding modes. In this embodiment, the corresponding mode between the calibration function and the node and the sub-node in the node is selected as the corresponding mode 1, and the first calibration graph is determined based on the corresponding mode 1. That is, one node corresponds to one calibration function, and the multiple sub-nodes in one node all correspond to the same calibration function, and at this time the node is also referred to as a calibration function node.

[0132] In an example, the calibration functions corresponding to the target qubits are determined; and a dependency relationship between the calibration functions is determined based on the calibration functions corresponding to the target qubits and a dependency between the target qubits. Optionally, the dependency between the target qubits can be represented by at least one of a dependency between related target qubits and a dependency between a group of parallelizable qubits. The related target qubits include at least one of repeatedly appearing target qubits, target qubits with a physical distance less than a threshold, and target qubits that affect each other. A group of qubits includes a plurality of parallelizable target qubits. In an example, the plurality of parallelizable target qubits can be at least one of target qubits of the same type and target qubits using the same calibration function. In the case where there is a dependency between the target qubits, there is a dependency relationship between the calibration functions corresponding to the target qubits.

[0133] In some embodiments, some restriction information can also be preset, for example, preset which calibration functions are conflicting calibration functions that cannot appear in the same calibration graph at the same time, or preset a second calibration function that needs to be executed before the first calibration function. When the dependency relationship between the calibration functions is determined based on the calibration functions corresponding to the target qubits and the dependency between the target qubits, the dependency relationship between the calibration functions is also optimized based on the restriction information, so that the optimized dependency relationship between the calibration functions can meet the above restriction information.

[0134] In step 540, the pre-configuration information is determined based on the dependency relationship between the calibration functions.

[0135] The pre-configuration information is some preset type of information used to determine the first calibration graph.

[0136] In some embodiments, the type of pre-configuration information includes at least one of node information (CalBatchSessionNode) and edge information (CalBatchSessionEdge). It should be noted that the node information refers to the information corresponding to each node in N nodes, which can also be referred to as a batch node (batch_node). A node can include a plurality of sub-nodes. Since a node can include a plurality of sub-nodes, the process of dividing the plurality of sub-nodes into a node can also be referred to as node pre-grouping.

[0137] Optionally, the node information comprises at least one of: a node identifier (batch_node_id) corresponding to the node, a calibration graph identifier (cal_session_id) corresponding to the node, a calibration function (functor) corresponding to the node, and at least one target qubit (ExecAssocQcompsGraph qcomps) corresponding to the calibration function. The at least one target qubit corresponding to the calibration function specifically comprises defining one calibration function as one node, defining single-qubit gates (single target qubit) and double-qubit gates (two adjacent qubits) in the target qubits corresponding to the calibration function as child nodes, and defining sub-edges between the child nodes based on the dependencies between the target qubits corresponding to the calibration function.

[0138] Optionally, the edge information comprises at least one of: a calibration graph identifier (cal_session_id) corresponding to the edge, an edge identifier (batch_edge_id) corresponding to the edge, and node identifiers of different nodes connected by the same edge. The different nodes connected by the same edge can be referred to as parent node and child node, and the node identifiers are parent node identifier (parent_node_id) and child node identifier (child_node_id). Optionally, in other embodiments, the edge information further comprises sub-edge information, and the sub-edge information comprises a calibration graph identifier corresponding to the sub-edge, a sub-edge identifier corresponding to the sub-edge, and sub-node identifiers of different sub-nodes connected by the same sub-edge. The above identifiers can be uniquely represented by one or more of letters, numbers, and characters.

[0139] For example, based on the dependencies between the calibration functions and the preset types, the pre-configuration information is determined. Specifically, a plurality of nodes corresponding to the calibration functions are determined; for each calibration function in the plurality of calibration functions, a plurality of child nodes are determined based on single-qubit gates and double-qubit gates in the target qubits corresponding to each calibration function, each child node in the plurality of child nodes corresponding to one target qubit of the single-qubit gate or two target qubits of the double-qubit gate; the dependencies between the nodes are determined based on the dependencies between the calibration functions; the dependencies between the plurality of child nodes in each node are determined based on the dependencies between the target qubits corresponding to each calibration function; the node information is determined based on the nodes and the child nodes, the edge information is determined based on the dependencies between the plurality of nodes and the dependencies between the child nodes in each node, and thus the pre-configuration information is obtained.

[0140] At step 560, based on the pre-configuration information, a first calibration graph is obtained; wherein the calibration function corresponds to at least one target qubit, and the calibration function is used to indicate a calibration manner of the at least one target qubit.

[0141] Exemplarily, the plurality of nodes and the plurality of sub-nodes in each node are plotted based on the node information in the pre-configuration information, and the edges between the plurality of nodes and the sub-edges between the plurality of sub-nodes in each node are plotted based on the edge information in the pre-configuration information, so as to determine the respective nodes, sub-nodes, edges between the nodes and sub-edges between the sub-nodes, and obtain the first calibration graph.

[0142] In the embodiment, a general manner of determining the first calibration graph is provided, the first calibration graph is determined by determining the dependency relationship between the calibration functions, and the dependency relationship between the calibration functions is determined based on the target qubits to be calibrated and the dependency between the target qubits, so that the dependency relationship between the calibration functions does not need to be manually specified, and the generation efficiency of the first calibration graph is improved.

[0143] Next, the quantum bit calibration method provided by the embodiments of the present application will be described in detail in combination with specific schematic diagrams, taking application to a quantum computer as an example.

[0144] ·Overall flow:

[0145] FIG. 6 is a schematic diagram of the quantum bit calibration method provided by one exemplary embodiment of the present application. The pre-configuration DAG graph (CalBatchSessionGraph) 30 is the first calibration graph as described above, and the calibration scheduling DAG graph (CalGraph) 50 is the second calibration graph as described above. Specifically, the quantum computer determines the calibration functions 10 corresponding to the target qubits, determines the dependency relationship 20 between the calibration functions, obtains the pre-configuration DAG graph 30, and optimizes the dependency relationship in the pre-configuration DAG graph 30 based on at least one expansion strategy in the automatic expansion strategy set (AutoStrategy) 40 and the pre-configuration DAG graph 30, to obtain the calibration scheduling DAG graph 50.

[0146] ·Definition example:

[0147] 1) Define the pre-configuration DAG graph prototype

[0148] For example, a preconfigured DAG (CalBatchSessionGraph) prototype, i.e. a standard template used when determining the first calibration graph, requires the following preconfigured information: node information (CalBatchSessionNode), edge information (CalBatchSessionEdge), calibration functions and their corresponding target qubits (ExecAssocQcompsGraph). The node information (CalBatchSessionNode) includes four types of information: batch node identification (batch_node_id), calibration graph identification (cal_session_id), calibration function (functor), and at least one target qubit corresponding to the calibration function (ExecAssocQcompsGraph qcomps). Specifically, a calibration function is defined as a node, and based on the target qubits corresponding to the calibration function, sub-nodes are determined, and based on the dependencies between the target qubits of the calibration function, sub-edges between multiple sub-nodes are determined. The edge information (CalBatchSessionEdge) includes four types of information: calibration graph identification (cal_session_id), edge identification (batch_edge_id) corresponding to the edge between batch nodes, and parent node identification (parent_node_id) and child node identification (child_node_id) connected by the same edge. Each of the above types of identification is represented by a 32-bit unsigned integer (unit32). It should be noted that the batch nodes in the preconfigured DAG graph correspond to the nodes in the first calibration graph, and the edges between batch nodes correspond to the edges between nodes in the first calibration graph.

[0149] 2) Defining a calibration scheduling DAG graph prototype

[0150] For example, a calibration scheduling DAG (CalNode) prototype, i.e. a template example of the second calibration graph, requires the following information: node information (CalNode), graph information (CalGraph). The node information (CalNode) includes four types of information: calibration graph identification (cal_session_id), node identification (cal_node_id), calibration function (functor), and target qubits. The graph information (CalGraph) includes edge information (cal_edge), which is represented by two nodes connected by the same edge. It should be noted that there are no batch nodes in the calibration scheduling DAG graph, and the nodes in the calibration scheduling DAG graph correspond to the sub-nodes of the target qubits with the finest granularity, i.e. the nodes in the calibration scheduling DAG graph correspond to the sub-nodes in the first calibration graph.

[0151] 3) Defining an automatic expansion strategy

[0152] Note: All auto-unfold strategies can be used in combination.

[0153] For example, the AutoStrategy set is used to determine at least one cross-node dependency. The AutoStrategy set includes three types of unfolding strategies: QcompsNameAnyMatch, QcompsNameAnyMatchRespectBatch, and PreserveBatchEdge. These unfolding strategies can be used individually or in combination.

[0154] • Calibration graph example:

[0155] • First calibration graph example:

[0156] The quantum computer determines the dependencies between the calibration functions based on the dependencies between the target qubits and the calibration functions corresponding to the target qubits, determines the subnodes corresponding to each calibration function and the dependencies between the subnodes, further determines a plurality of subnodes that can be batched to determine a node, and determines the dependencies between the nodes to obtain a first calibration graph.

[0157] Figure 7 is a schematic diagram of a first calibration graph according to an example embodiment of the present application. The first calibration graph includes a plurality of nodes, each node includes a plurality of subnodes, each calibration function corresponds to a node, each calibration function corresponds to one or more target qubits, the target qubits are represented by the letters q and numbers, and the nodes and subnodes are represented by numbers.

[0158] Specifically, the first calibration graph includes four nodes: node 1, node 2, node 3, and node 4. The dependencies between the four nodes are represented by edges, which are directed edges. Node 1 points to node 2 and node 3, node 2 points to node 4, and node 3 points to node 4.

[0159] For node 1, node 1 includes five subnodes: subnode 1, subnode 2, subnode 3, subnode 4, and subnode 5. The dependencies between the five subnodes are represented by subedges, which are directed edges. Subnode 1 points to subnode 3, which points to subnode 5, subnode 2 points to subnode 4, which points to subnode 5. Subnode 1 corresponds to q1, subnode 2 corresponds to q2, subnode 3 corresponds to q3, subnode 4 corresponds to q4, and subnode 5 corresponds to q4 and q5.

[0160] For node 2, node 2 includes 4 child nodes: child node 6, child node 7, child node 8 and child node 9. The dependency relationship between the 4 child nodes is represented by child edges, which are directed edges, child node 6 pointing to child node 8 pointing to child node 9, child node 7 pointing to child node 8 pointing to child node 9. Child node 6 corresponds to q4 and q5, child node 7 corresponds to q7, child node 8 corresponds to q8, and child node 9 corresponds to q9.

[0161] For node 3, node 3 includes 4 child nodes: child node 10, child node 11, child node 12 and child node 13. The dependency relationship between the 4 child nodes is represented by child edges, which are directed edges, child node 10 pointing to child node 12, child node 11 pointing to child node 13. Child node 10 corresponds to q3, child node 11 corresponds to q5, child node 12 corresponds to q10, and child node 13 corresponds to q11.

[0162] For node 4, node 4 includes 3 child nodes: child node 14, child node 15 and child node 16. The dependency relationship between the 3 child nodes is represented by child edges, which are directed edges, child node 14 pointing to child node 15 pointing to child node 16. Child node 14 corresponds to q9, child node 15 corresponds to q10 and q11, and child node 16 corresponds to q12.

[0163] • Second calibration graph example 1, target qubit dependency expansion (QcompsNameAnyMatch):

[0164] The target qubit dependency expansion (QcompsNameAnyMatch) is used to determine the first dependency relationship. The quantum computer determines the first dependency relationship between each child node in the first node and each child node in the second node based on the dependency between the target qubits corresponding to each child node in the first node and the target qubits corresponding to each child node in the second node. Add new edges corresponding to the first dependency relationship in the first calibration graph to obtain the second calibration graph. Wherein, the first node and the second node are any two nodes in the first calibration graph.

[0165] FIG. 8 is a schematic diagram of a second calibration graph according to an example embodiment of the present application. Compared with FIG. 7, a total of 7 new edges corresponding to the first dependency relationship are added in the first calibration graph, which are as follows:

[0166] 1) The first new edge: the child edge between child node 5 and child node 6: cal_edge(calnode5, calnode6), calnode5(q4, q5), calnode6(q4, q5) / / q4 and q5 have a dependency relationship;

[0167] 2) New edge: Sub-edge between child node 5 and child node 11: cal_edge(calnode5, calnode11), calnode5(q4, q5), calnode11(q5) / / q5 exists dependency;

[0168] 3) New edge: Sub-edge between child node 9 and child node 14: cal_edge(calnode9, calnode14), calnode9(q9), calnode14(q9) / / q9 exists dependency;

[0169] 4) New edge: Sub-edge between child node 12 and child node 15: cal_edge(calnode12, calnode15), calnode12(q10), calnode15(q10, q11) / / q10 exists dependency;

[0170] 5) New edge: Sub-edge between child node 13 and child node 15: cal_edge(calnode13, calnode15), calnode13(q11), calnode15(q10, q11) / / q11 exists dependency;

[0171] 6) New edge: Sub-edge between child node 3 and child node 10: cal_edge(calnode3, calnode10), calnode3(q3), calnode10(q3) / / q3 exists dependency;

[0172] 7) New edge: Sub-edge between child node 4 and child node 6: cal_edge(calnode4, calnode6), calnode4(q4), calnode6(q4, q5) / / q4 exists dependency.

[0173] • Second calibration graph example 2, respecting node pre-grouping target qubit dependency unfolding (QcompsNameAnyMatchRespectBatch):

[0174] The QcompsNameAnyMatchRespectBatch is used for determining the second dependency, which is a subset of the first dependency. The quantum computer determines the end nodes in the first node from the respective sub-nodes in the first node; and determines the second dependency between the end nodes in the first node and the respective sub-nodes in the second node based on the dependency between the target qubits corresponding to the end nodes in the first node and the target qubits corresponding to the respective sub-nodes in the second node. The second dependency is added in the first calibration graph to obtain a second calibration graph. The first node and the second node are two nodes in the first calibration graph that have a dependency between nodes.

[0175] FIG. 9 is a schematic diagram of a second calibration graph according to an example embodiment of the present application. Compared with FIG. 7, five new edges corresponding to the second dependencies are added in the first calibration graph, which are as follows:

[0176] 1) The first new edge: the sub-edge between the sub-node 5 and the sub-node 6: cal_edge(calnode5, calnode6), calnode5(q4, q5), calnode6(q4, q5) / / q4 and q5 have a dependency;

[0177] 2) The second new edge: the sub-edge between the sub-node 5 and the sub-node 11: cal_edge(calnode5, calnode11), calnode5(q4, q5), calnode11(q5) / / q5 has a dependency;

[0178] 3) The third new edge: the sub-edge between the sub-node 9 and the sub-node 14: cal_edge(calnode9, calnode14), calnode9(q9), calnode14(q9) / / q9 has a dependency;

[0179] 4) The fourth new edge: the sub-edge between the sub-node 12 and the sub-node 15: cal_edge(calnode12, calnode15), calnode12(q10), calnode15(q10, q11) / / q10 has a dependency;

[0180] 5) The fifth new edge: the sub-edge between the sub-node 13 and the sub-node 15: cal_edge(calnode13, calnode15), calnode13(q11), calnode15(q10, q11) / / q11 has a dependency.

[0181] • The second calibration graph example 3, node pre-batch edge preserving:

[0182] The node pre-batch edge preserving (PreserveBatchEdge) is used to determine the third dependency. The quantum computer determines, from each of the child nodes in the first node, a terminal child node in the first node; and determines, from each of the child nodes in the second node, a starting child node in the second node; determines a third dependency between the terminal child node in the first node and the starting child node in the second node. The second calibration graph is obtained by adding the new edges corresponding to the second dependency in the first calibration graph. Wherein, the first node and the second node are two nodes in the first calibration graph that exist between the nodes dependency relationship.

[0183] Figure 10 is a schematic diagram of the second calibration graph provided by one of the exemplary embodiments of the present application. Compared with Figure 7, a total of 7 new edges corresponding to the third dependency are added in the first calibration graph, which are as follows:

[0184] 1) The first new edge: the sub-edge between the child node 5 and the child node 6: cal_edge(calnode5, calnode6), calnode5(q4, q5), calnode6(q4, q5);

[0185] 2) The second new edge: the sub-edge between the child node 5 and the child node 11: cal_edge(calnode5, calnode11), calnode5(q4, q5), calnode11(q5);

[0186] 3) The third new edge: the sub-edge between the child node 5 and the child node 7: cal_edge(calnode5, calnode7), calnode5(q4, q5), calnode7(q7);

[0187] 4) The fourth new edge: the sub-edge between the child node 5 and the child node 10: cal_edge(calnode5, calnode10), calnode5(q4, q5), calnode10(q3);

[0188] 5) The fifth new edge: the sub-edge between the child node 9 and the child node 14: cal_edge(calnode9, calnode14), calnode9(q9), calnode14(q9);

[0189] 6) The 6th newly added edge: the sub-edge between the child node 12 and the child node 14: cal_edge(calnode12, calnode14), calnode12(q10), calnode14(q9);

[0190] 7) The 7th newly added edge: the sub-edge between the child node 13 and the child node 14: cal_edge(calnode13, calnode14), calnode13(q11), calnode14(q9).

[0191] • The second calibration graph example 4, the target qubit dependency expansion (QcompsNameAnyMatch) and the node pre-grouping dependency expansion (PreserveBatchEdge):

[0192] The quantum computer superposition uses the target qubit dependency expansion (QcompsNameAnyMatch) and the node pre-grouping dependency expansion (PreserveBatchEdge), while preserving the dependency relationships determined by the two methods. Add the newly added edges corresponding to the two dependency relationships in the first calibration graph to obtain the second calibration graph. It should be noted that the present embodiment is only an example, and can also be some other superposition combinations.

[0193] FIG. 11 is a schematic diagram of a second calibration graph according to an example embodiment of the present application. Compared with FIG. 7, a total of 11 newly added edges corresponding to the dependency relationships are added in the first calibration graph, which are as follows:

[0194] 1) The 1st newly added edge: the sub-edge between the child node 5 and the child node 6: cal_edge(calnode5, calnode6), calnode5(q4, q5), calnode6(q4, q5) / / q4, q5 exist dependency;

[0195] 2) The 2nd newly added edge: the sub-edge between the child node 5 and the child node 11: cal_edge(calnode5, calnode11), calnode5(q4, q5), calnode11(q5) / / q5 exists dependency;

[0196] 3) The 3rd newly added edge: the sub-edge between the child node 5 and the child node 7: cal_edge(calnode5, calnode7), calnode5(q4, q5), calnode7(q7);

[0197] 4) The 4th new edge: the child edge between child node 5 and child node 10: cal_edge(calnode5, calnode10), calnode5(q4, q5), calnode10(q3);

[0198] 5) The 5th new edge: the child edge between child node 9 and child node 14: cal_edge(calnode9, calnode14), calnode9(q9), calnode14(q9);

[0199] 6) The 6th new edge: the child edge between child node 12 and child node 14: cal_edge(calnode12, calnode14), calnode12(q10), calnode14(q9);

[0200] 7) The 7th new edge: the child edge between child node 13 and child node 14: cal_edge(calnode13, calnode14), calnode13(q11), calnode14(q9);

[0201] 8) The 8th new edge: the child edge between child node 12 and child node 15: cal_edge(calnode12, calnode15), calnode12(q10), calnode15(q10, q11) / / q10 exists dependency;

[0202] 9) The 9th new edge: the child edge between child node 13 and child node 15: cal_edge(calnode13, calnode15), calnode13(q11), calnode15(q10, q11) / / q11 exists dependency;

[0203] 10) The 10th new edge: the child edge between child node 3 and child node 10: cal_edge(calnode3, calnode10), calnode3(q3), calnode10(q3) / / q3 exists dependency;

[0204] 11) The 11th new edge: the child edge between child node 4 and child node 6: cal_edge(calnode4, calnode6), calnode4(q4), calnode6(q4, q5) / / q4 exists dependency.

[0205] • Application scenarios:

[0206] The calibration method of the quantum bit provided in the embodiments of the present application can realize the definition of the calibration function node in the parameter reference mode, and complete the automatic construction of the second calibration graph according to the first calibration graph. Subsequently, the calibration of the quantum bit can be realized based on the second calibration graph, and the research and development efficiency of calibration automation is improved.

[0207] In summary, the calibration method of the quantum bit provided in the embodiments of the present application solves the problem of excessively high cost of manually defining the calibration graph in the calibration automation system construction process of the quantum chip, and the entire process does not require human intervention, thereby improving the calibration efficiency of the quantum bit. Specifically, by defining a universal and standard first calibration graph and optimizing the first calibration graph, the first calibration graph is automatically converted into a second calibration graph, thereby reducing the labor cost of manually defining the calibration graph. At the same time, by designing a diversified and expandable automatic expansion strategy set from the first calibration graph to the second calibration graph, the openness and freedom of generating the second calibration graph can be retained, the construction cost of automatic calibration is greatly reduced, the conversion efficiency is improved, and the calibration efficiency of the quantum bit is greatly improved.

[0208] FIG. 12 is a block diagram of a quantum bit calibration apparatus 800 provided in an example embodiment of the present application. The quantum bit calibration apparatus 800 includes at least part of the following modules: an acquisition module 810, an optimization module 820, and a calibration module 830.

[0209] The acquisition module 810 is configured to acquire a first calibration graph, the first calibration graph including N nodes and edges between the N nodes, the edges being used to represent a dependency relationship between the N nodes, each of the N nodes being used to represent a calibration process for at least one target quantum bit, and the dependency relationship in the first calibration graph being determined based on a part of dependency of the target quantum bits corresponding to the N nodes, N being a positive integer.

[0210] The optimization module 820 is configured to optimize the dependency relationship in the first calibration graph to obtain a second calibration graph, the completeness of the dependency relationship in the second calibration graph being higher than the completeness of the dependency relationship in the first calibration graph.

[0211] The calibration module 830 is configured to calibrate the target quantum bit based on the second calibration graph.

[0212] In some embodiments, the optimization module 820 is configured to:

[0213] add at least one new edge in the first calibration graph to obtain the second calibration graph;

[0214] The at least one added edge corresponds to one-to-one mapping of at least one cross-node dependency relationship, and the at least one cross-node dependency relationship includes a dependency relationship between two sub-nodes belonging to different nodes in the N nodes.

[0215] In some embodiments, each node in the N nodes corresponds to the at least one target qubit; an optimization module 820 is configured to:

[0216] Based on at least one of an inter-node dependency relationship, an intra-node dependency relationship, and a target qubit corresponding to a sub-node in different nodes in the N nodes in the first calibration graph, the at least one added edge is added in the first calibration graph to obtain the second calibration graph.

[0217] The inter-node dependency relationship is a dependency relationship between different nodes in the N nodes, and the intra-node dependency relationship is a dependency relationship between different sub-nodes in the same node in the N nodes.

[0218] In some embodiments, the at least one cross-node dependency relationship includes a first dependency relationship, the first dependency relationship is a dependency relationship between each sub-node in a first node and each sub-node in a second node, and the first node and the second node are different nodes in the N nodes.

[0219] In some embodiments, the optimization module 820 is further configured to:

[0220] Based on a dependency between a target qubit corresponding to each sub-node in the first node and a target qubit corresponding to each sub-node in the second node, the first dependency relationship between each sub-node in the first node and each sub-node in the second node is determined.

[0221] In some embodiments, the at least one cross-node dependency relationship includes a second dependency relationship, the second dependency relationship is a subset of the first dependency relationship, and the second dependency relationship is a dependency relationship between a terminal sub-node in the first node and each sub-node in the second node, and the first node and the second node are different nodes in the N nodes in which the inter-node dependency relationship exists.

[0222] In some embodiments, the optimization module 820 is further configured to:

[0223] Based on the intra-node dependency relationship of the first node, the terminal sub-node in the first node is determined from each sub-node in the first node.

[0224] determine the second dependency relationship between the terminal sub-node in the first node and each sub-node in the second node based on dependencies between target qubits corresponding to the terminal sub-node in the first node and target qubits corresponding to each sub-node in the second node.

[0225] In some embodiments, the at least one cross-node dependency relationship includes a third dependency relationship between a terminal sub-node in the first node and a starting sub-node in the second node, the first node and the second node being different nodes in the N nodes in which the inter-node dependency relationship exists.

[0226] In some embodiments, the optimization module 820 is further configured to:

[0227] determine the terminal sub-node in the first node from each sub-node in the first node based on the intra-node dependency relationship of the first node, and determine the starting sub-node in the second node from each sub-node in the second node based on the intra-node dependency relationship of the second node;

[0228] determine the third dependency relationship between the terminal sub-node in the first node and the starting sub-node in the second node.

[0229] In some embodiments, the obtaining module 810 is configured to:

[0230] determine dependency relationships between the calibration functions based on the dependencies between the target qubits and the calibration functions corresponding to the target qubits;

[0231] determine pre-configuration information based on the dependency relationships between the calibration functions;

[0232] obtain the first calibration graph based on the pre-configuration information;

[0233] The calibration function corresponds to at least one target qubit, and the calibration function is used to indicate a calibration manner of the at least one target qubit.

[0234] In some embodiments, the type of the pre-configuration information includes at least one of node information and edge information;

[0235] The node information includes at least one of the following: a node identifier corresponding to a node, a calibration graph identifier corresponding to the node, a calibration function corresponding to the node, and at least one target qubit corresponding to the calibration function.

[0236] The side information includes at least one of the following: the calibration graph identifier corresponding to the side, the side identifier corresponding to the side, and the node identifier of different nodes connected by the same side.

[0237] It should be noted that the specific limitations in the above-mentioned embodiments of the calibration device 800 of one or more qubits can refer to the limitations of the calibration method of the qubit described above, which will not be repeated here. The modules of the above device can be realized by software, hardware and their combination, and each module can be embedded in the form of hardware or independent of the processor of the quantum computer, or stored in the memory of the quantum computer in the form of software, so that the processor calls and executes the corresponding operations of each module.

[0238] The embodiments of the present application also provide a quantum computer, which can be implemented by a classical computer, and the quantum computer comprises a processor and a memory, the memory stores a computer program; the processor is configured to execute the computer program in the memory to implement the calibration method of the qubit provided by the above-mentioned method embodiments.

[0239] FIG. 13 is a structural block diagram of a quantum computer 1000 according to an example embodiment of the present application.

[0240] Generally, the quantum computer 1000 comprises a processor 1001 and a memory 1002.

[0241] The processor 1001 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 1001 can be implemented in at least one of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 1001 can also include a main processor and a coprocessor. The main processor is a processor for processing data in the wake-up state, also known as a central processing unit (CPU). The coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 1001 can be integrated with a graphics processing unit (GPU), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 1001 can also include an artificial intelligence (AI) processor, which is used to process machine learning-related computing operations.

[0242] The memory 1002 can include one or more computer-readable storage media. The computer-readable storage media can be non-transitory. The memory 1002 can also include high-speed random access memory and can include non-volatile memory, such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile storage devices. In some embodiments, the non-transitory computer-readable storage media of the memory 1002 is used to store at least one instruction for execution by the processor 1001 to implement the method for calibrating a quantum bit provided by the method embodiments.

[0243] In some embodiments, the quantum computer 1000 can also optionally include an input interface 1003 and an output interface 1004. The processor 1001, the memory 1002, and the input interface 1003 and the output interface 1004 can be connected by a bus or a signal line. Various peripheral devices can be connected to the input interface 1003 and the output interface 1004 through the bus, the signal line, or the circuit board. The input interface 1003 and the output interface 1004 can be used to connect at least one peripheral device related to input / output (I / O) to the processor 1001 and the memory 1002. In some embodiments, the processor 1001, the memory 1002, and the input interface 1003 and the output interface 1004 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 1001, the memory 1002, and the input interface 1003 and the output interface 1004 can be implemented on a separate chip or circuit board, and the embodiments of the present application do not limit this.

[0244] The embodiments of the present application also provide another quantum computer, which can be implemented by at least one of a software implementation architecture and a hardware implementation architecture. The software implementation architecture is used to implement the method for calibrating a quantum bit provided by the above method embodiments. The hardware implementation architecture is used to provide hardware support for the software implementation architecture, and is used to support the software implementation architecture to implement the method for calibrating a quantum bit provided by the above method embodiments.

[0245] FIG. 14 is a structural block diagram of a quantum computer 1200 provided by an example embodiment of the present application. The quantum computer 1200 can be implemented by a hardware implementation architecture 1210 and / or a software implementation architecture 1220.

[0246] For example, the hardware implementation architecture 1210 is constructed based on a quantum bit network. The quantum bit is the basic unit of quantum information. A plurality of quantum bits are connected to each other through a quantum circuit, which is a set of quantum logic gates used to control the flow of information between quantum bits. The quantum circuit also needs to be connected to a classical computer for controlling the quantum circuit and reading the calculation results.

[0247] Specifically, the hardware implementation architecture 1210 includes: a quantum bit unit (Qubits) 1211, a quantum controller (Quantum Controller) 1212, a quantum entanglement source (Quantum Entanglement Source) 1213, a quantum operator (Quantum Operator) 1214, and a quantum detector (Quantum Detector) 1215. Among them, the quantum bit unit 1211 is the basic unit of the quantum computer 1200, which can represent two states: 0 and 1 at the same time. The quantum controller 1212 is used to control the state of the quantum bit and the interaction between the quantum bits. The quantum entanglement source 1213 is used to generate quantum entanglement, which can make the quantum bits entangled, thereby improving the performance of the quantum computer 1200. The quantum operator 1214 is used to operate the quantum bit. The quantum detector 1215 is used to detect the state of the quantum bit.

[0248] For example, the software implementation architecture 1220 includes: a quantum processor 1227, a classical processor 1226, and a quantum programming language 1221. The quantum processor 1227 is used to execute the quantum program to implement the quantum bit calibration method provided by the above-mentioned method embodiments; the classical processor 1226 is used to manage the quantum processor 1227 and provide an interface for the user; the quantum programming language 1221 is a high-level programming language for the quantum program executed on the quantum processor 1227, including at least one of Q#, Qiskit, and OpenQASM.

[0249] Optionally, the software implementation architecture 1220 further includes a quantum compiler 1222, a quantum simulator 1223, a quantum debugger 1224, and a quantum library 1225. Specifically, the quantum compiler 1222 is used to compile the quantum program into quantum instructions for execution on the quantum computer 1200. The quantum simulator 1223 is a software used to simulate the quantum program, which can simulate the execution process of the quantum program and the execution result of the quantum program. The quantum debugger 1224 is a software used to debug the quantum program, which can check the correctness of the quantum program and the performance of the quantum program. The quantum library 1225 is used to provide reusable components of the quantum program, which can improve the development efficiency of the quantum program.

[0250] Those skilled in the art can understand that the structure shown in FIG. 13 does not constitute a limitation on the quantum computer 1000, and the structure shown in FIG. 14 does not constitute a limitation on the quantum computer 1200, which can include more or fewer components than those shown, or combine certain components, or adopt different component arrangements.

[0251] In an example embodiment, the application provides a quantum chip, the quantum chip comprising a programmable logic circuit and / or program instructions, when the quantum chip is running on a quantum computer, for implementing the calibration method of the qubit provided by the above method embodiments.

[0252] The application provides a computer readable storage medium, the computer readable storage medium storing a computer program, the computer program being loaded and executed by a processor to implement the calibration method of the qubit provided by the above method embodiments.

[0253] The application provides a computer program product or computer program, which can be implemented as a quantum program. The computer program product or computer program comprises computer instructions stored in a computer readable storage medium. The processor of the quantum computer reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the processor of the quantum computer is loaded and executed to implement the calibration method of the qubit provided by the above method embodiments.

[0254] The above application embodiment numbers are only for description, not representing the advantages and disadvantages of the embodiments.

[0255] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by program instructing relevant hardware to complete, and the program can be stored in a computer readable storage medium. The computer readable storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk.

[0256] Those skilled in the art should realize that in the above one or more examples, the functions described in the embodiments of the application can be realized by hardware, software, firmware or any combination thereof. When realized by software, these functions can be stored in a computer readable medium or transmitted as one or more instructions or codes on a computer readable medium. The computer readable medium includes computer storage medium and communication medium, wherein the communication medium includes any medium facilitating the transmission of computer programs from one place to another. The storage medium can be any available medium accessible by a general or special purpose computer.

[0257] The above is only an optional embodiment of the application, and does not limit the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the protection scope of the application.

Claims

1. A method for calibrating a quantum bit, the method being performed by a quantum computer, the method comprising: obtaining a first calibration graph, the first calibration graph comprising N nodes and edges between the N nodes, the edges being used to represent dependencies between the N nodes, each of the N nodes being used to represent a calibration process for at least one target quantum bit, the dependencies in the first calibration graph being determined based on a portion of dependencies of the target quantum bits corresponding to the N nodes, N being a positive integer; optimizing the dependencies in the first calibration graph to obtain a second calibration graph, the dependencies in the second calibration graph having a higher completeness than the dependencies in the first calibration graph; and calibrating the target quantum bits based on the second calibration graph. The optimizing the dependencies in the first calibration graph to obtain a second calibration graph comprises: adding at least one new edge in the first calibration graph to obtain the second calibration graph; wherein the at least one new edge corresponds to at least one cross-node dependency, the at least one cross-node dependency comprising a dependency between two sub-nodes belonging to different nodes in the N nodes. Each of the N nodes corresponds to the at least one target quantum bit. The adding at least one new edge in the first calibration graph to obtain a second calibration graph comprises: adding the at least one new edge in the first calibration graph based on at least one of inter-node dependencies in the first calibration graph, intra-node dependencies in the first calibration graph, and target quantum bits corresponding to sub-nodes in different nodes in the N nodes, to obtain the second calibration graph; wherein the inter-node dependencies are dependencies between different nodes in the N nodes, and the intra-node dependencies are dependencies between different sub-nodes in a same node in the N nodes.

2. The method of claim 1, wherein, The at least one cross-node dependency comprises a first dependency, the first dependency being a dependency between each sub-node in a first node and each sub-node in a second node, the first node and the second node being different nodes in the N nodes; and the method further comprises: determining the first dependency between each sub-node in the first node and each sub-node in the second node based on a dependency between target quantum bits corresponding to each sub-node in the first node and target quantum bits corresponding to each sub-node in the second node. The at least one cross-node dependency comprises a second dependency, the second dependency being a subset of the first dependency, the second dependency being a dependency between a terminal sub-node in the first node and each sub-node in the second node, the first node and the second node being different nodes in the N nodes having the inter-node dependencies. The method further comprises: determining the terminal sub-node in the first node from each sub-node in the first node based on the intra-node dependencies of the first node.

3. The method of claim 2, wherein, ​ ​ ​ ​ 4. The method of claim 3, wherein, ​ ​ 5. The method of claim 4, wherein, ​ ​ ​ determine the second dependency relationship between the terminal sub-node in the first node and each sub-node in the second node based on dependencies between target qubits corresponding to the terminal sub-node in the first node and target qubits corresponding to each sub-node in the second node.

6. The method of claim 3, wherein, The at least one cross-node dependency relationship includes a third dependency relationship between a terminal sub-node in the first node and a starting sub-node in the second node, the first node and the second node being different nodes in the N nodes in which the inter-node dependency relationship exists; the method further comprises: determining the terminal sub-node in the first node from each sub-node in the first node based on the intra-node dependency relationship of the first node, and determining the starting sub-node in the second node from each sub-node in the second node based on the intra-node dependency relationship of the second node; determining the third dependency relationship between the terminal sub-node in the first node and the starting sub-node in the second node.

7. The method according to any one of claims 1 to 6, wherein, The first calibration graph includes: determining dependency relationships between the calibration functions based on the calibration functions corresponding to the target qubits and the dependencies between the target qubits; determining pre-configuration information based on the dependency relationships between the calibration functions; obtaining the first calibration graph based on the pre-configuration information; The calibration function corresponds to at least one target qubit, and the calibration function is used to indicate a calibration manner of the at least one target qubit.

8. The method of claim 7, wherein, The type of the pre-configuration information includes at least one of node information and edge information; The node information includes at least one of the following: a node identifier corresponding to a node, a calibration graph identifier corresponding to the node, a calibration function corresponding to the node, and at least one target qubit corresponding to the calibration function; The edge information includes at least one of the following: a calibration graph identifier corresponding to an edge, an edge identifier corresponding to the edge, and node identifiers of different nodes connected by the same edge.

9. A quantum bit calibration device, the device comprising: an acquisition module configured to acquire a first calibration graph, the first calibration graph including N nodes and edges between the N nodes, the edges being used to represent dependency relationships between the N nodes, each node in the N nodes being used to represent a calibration process for at least one target qubit, and dependency relationships in the first calibration graph being determined based on a part of dependencies of target qubits corresponding to the N nodes, N being a positive integer; an optimization module configured to optimize the dependency relationships in the first calibration graph to obtain a second calibration graph, the completeness of dependency relationships in the second calibration graph being higher than the completeness of dependency relationships in the first calibration graph; a calibration module configured to calibrate the target qubits based on the second calibration graph.

10. A quantum chip comprising programmable logic circuitry or program instructions, for implementing the method of calibrating a qubit according to any one of claims 1 to 8 when the quantum chip is run on a quantum computer.

11. A quantum computer comprising: A processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the method of calibrating a qubit according to any one of claims 1 to 8.

12. A computer readable storage medium storing a computer program, the computer program being loaded and executed by a processor to implement the method of calibrating a qubit according to any one of claims 1 to 8.

13. A computer program product comprising computer instructions stored in a computer readable storage medium, the computer instructions being fetched by a processor from the computer readable storage medium, causing the processor to load and execute to implement the method of calibrating a qubit according to any one of claims 1 to 8.

Images