Reducing the impact of noise on quantum state identification through phase shifting.

Abstract

A method, system, and computer program product for improving the identification of qubit states in a superconducting quantum computer. A channel is calibrated to train a kernel to contain accurate calibration data ("kernel states"). After calibrating the channel, a test is performed in which quantum operations are simultaneously performed on qubits in adjacent channels, including the recently calibrated channel, to determine whether the kernel response differs from the expected kernel response, where the expected kernel response exceeds a threshold based on the kernel states of the trained kernel of the recently calibrated channel. If such a situation occurs, the phase of the signal for the recently calibrated channel is shifted, and the process of recalibrating and testing the channel (using the phase-shifted signal) is repeated until the difference between the kernel response and the expected kernel response no longer exceeds a threshold.

Description

Technical Field

[0001] The present disclosure generally relates to superconducting quantum computing, and more specifically, to improving the discrimination of qubit states (e.g., transmon qubit states) in a superconducting quantum computer by phase-shifting the read signal to change the impact the integrated result of the qubit kernel filter.

Background Art

[0002] Superconducting quantum computing is a field of solid-state quantum computing that realizes superconducting electronic circuits using superconducting qubits as artificial atoms or quantum dots. In superconducting qubits, the two logical states are the ground state and the excited state, represented by |g> and |e>, or |0> and |1>, respectively.

[0003] Superconductors are implemented due to the fact that they have almost infinite conductivity and almost zero resistance at low temperatures. Each qubit is constructed using a semiconductor circuit including Josephson junctions and capacitors.

[0004] Superconducting capacitors and Josephson junctions are used to generate a resonant circuit that almost does not release energy as heat that may interfere with quantum information. The superconducting resonant circuit is a type of artificial atom that can be used as a qubit.

[0005] A qubit is a generalization of a bit (a system with two possible states) that can take the quantum superposition of both states. On the other hand, a quantum gate is a generalization of logic gates that describe the transformations that occur when a gate is applied to one or more qubits, given their initial states. The physical implementation of qubits and gates is difficult for the same reasons that quantum phenomena are difficult to observe in everyday life, given the extremely small scales on which they occur. One approach to realizing a quantum computer is to implement superconductors, which would make quantum effects macroscopically observable, but at the cost of requiring extremely low operating temperatures.

[0006] A special type of qubit known as a transmon may be used in such superconducting quantum computers. A transmon is a special type of qubit with a shunt capacitor specifically designed to reduce noise. The transmon qubit model is based on Cooper pairs boxes and reduces noise by increasing the ratio of Josephson energy to charge energy.

[0007] In a superconducting quantum computer, such a transmon qubit is controlled via the use of microwave radio frequency signals; that is, such a transmon qubit can be manipulated by microwave radio frequency signals. The state of such a transmon qubit is then read, for example, via the use of a microwave resonator, where the resonant frequency of the resonator is shifted dispersively by the qubit state.

[0008] As the number of qubits increases, the useful frequency range becomes crowded and noisy, and effective computation diminishes. In particular, identifying transmon qubit states becomes difficult, especially when using dynamic circuits in kernel threshold systems.

[0009] A dynamic circuit is a quantum circuit that includes an intermediate circuit measurement, also known as a feedforward operation, which affects the control flow of subsequent gate executions in the circuit. The standard and simplest example is a conditional reset, where a projection measurement of a qubit is performed. If the result is |1>, the state is flipped from the |1> state to the |0> state using the X gate. If the result is |0>, the state is not flipped.

[0010] The dynamic circuit utilizes a low-latency decision architecture known as a "kernel," which performs evaluations based on the state read from the qubit. The "kernel" refers to a time-domain filter calibrated for a specific task. Next, real-time convolution of the arriving signal (e.g., a microwave radio frequency signal) is performed on the kernel to generate a combined result (also called the "kernel response") that is compared to a threshold to determine the quantum state of the qubit (e.g., a transmon qubit). For example, if the combined result exceeds the threshold, the result may be considered 1, and if the combined result is below the threshold, the result may be considered 0.

[0011] When calculations involving such dynamic circuits are extended to other states, such as |2> and |3> states, in addition to |0> and |1> states, it becomes difficult to maintain the low computational latency required to enable proper operation. As a result, a fast decision-making kernel is applied to sampled data in real time. Such outputs are compared to a set of thresholds to determine the state readout.

[0012] Unfortunately, kernel thresholding systems require a sufficiently large kernel sample size to clearly distinguish between input states. For example, the sample size can range from thousands to tens of thousands or even more. In latency-sensitive applications, such large sample sizes can result in long capture windows, thereby slowing down the responsiveness of dynamic circuits.

[0013] Furthermore, kernel threshold systems are often susceptible to in-band noise (e.g., crosstalk, Nyquist aliasing). Crosstalk is a disturbance caused by the electric or magnetic field of a signal affecting an adjacent signal. Nyquist aliasing refers to noise and signal images that are folded back from a higher Nyquist band to a first Nyquist band. As a result of such in-band noise, more samples are required, thereby further increasing the capture window and further slowing down the ability of the dynamic circuit to respond. Moreover, such noise can consequently make it difficult to identify qubit states (e.g., transmon qubit states).

[0014] In addition, in multiplexed applications, bleed-over energy (energy spread across adjacent channels) can introduce dynamic decision-making problems in a similar manner to crosstalk, including difficulties in identifying qubit states (e.g., transmonqubit states).

[0015] As a result, there is considerable interest in solutions to address the difficulty of identifying qubit states (e.g., transmonqubit states) in superconducting quantum computers, particularly those using dynamic circuits in kernel threshold systems. [Overview of the project]

[0016] In one embodiment of the present disclosure, a method for improving the identification of qubit states in a superconducting quantum computer comprises the step of shifting the phase of a signal in response to a kernel response not meeting a criterion. The method further comprises the step of using the phase-shifted signal to measure the state of a first qubit in response to the kernel response meeting the criterion.

[0017] In addition, in one embodiment of the present disclosure, the method further comprises a step of continuing to adjust the phase of the phase-shifted signal until the kernel response satisfies the criterion.

[0018] Furthermore, in one embodiment of the present disclosure, the method additionally includes a step of calibrating each of the multiple channels used to communicate the measured state of the qubit.

[0019] In addition, in one embodiment of the present disclosure, the method further comprises reading a plurality of responses of the second qubit set to the first quantum state for a first channel of the plurality of channels, as a result of a dynamic circuit performing a quantum operation on the second qubit set to the first quantum state.

[0020] Furthermore, in one embodiment of the present disclosure, the method further comprises the step of generating and recording kernel states for the first and second quantum states and storing them in a kernel based on the read responses of the second qubit set to the first and second quantum states in response to the completion of calibration of the first channel, wherein the kernel response is obtained from the calibrated first channel using the kernel states.

[0021] In addition, in one embodiment of the present disclosure, the method comprises the step of setting the excited state of a third qubit to a certain quantum state. The method further comprises the step of reading the response of the third qubit as a result of a dynamic circuit performing a quantum operation on the third qubit set to the certain quantum state. Furthermore, the method comprises the step of determining the kernel response of the third qubit based on the read response of the third qubit and the kernel state. In addition, the method comprises the step of comparing the kernel response of the third qubit with the kernel response obtained from the calibrated first channel. Furthermore, the method comprises the step of shifting the phase of the signal of the calibrated first channel in response to the difference between the kernel response of the third qubit and the kernel response obtained from the calibrated first channel exceeding a threshold.

[0022] Furthermore, in one embodiment of the present disclosure, the method comprises the step of recalibrating the first channel using the phase-shifted signal.

[0023] Other embodiments of the methods described above include systems and computer program products.

[0024] Accordingly, embodiments of the present disclosure improve the identification of qubit states in a superconducting quantum computer by shifting the phase of the signal used to measure the state of a qubit, thereby eliminating the effects of noise when identifying qubit states and increasing the speed and accuracy of state identification.

[0025] The above provides a fairly general overview of the features and technical advantages of one or more embodiments of the present disclosure in order to better understand the detailed description of the present disclosure below. Additional features and advantages of the present disclosure will be described hereafter and may form the subject matter of the claims of the present disclosure. [Brief explanation of the drawing]

[0026] A better understanding of the present disclosure can be obtained when the following detailed description is considered in conjunction with the following drawings.

[0027] [Figure 1] A communication system for practicing the principles of the present disclosure according to an embodiment of the present disclosure is shown.

[0028] [Figure 2] A kernel response is shown from the perspective of the calculation range for the readout threshold with respect to the frequency and phase of a signal according to an embodiment of the present disclosure.

[0029] [Figure 3] A diagram of a software component of a classical computer used to improve the discrimination of qubit states in a superconducting quantum computer by shifting the phase of a signal used to measure the state of a qubit, thereby removing the influence of noise when discriminating the qubit state, according to an embodiment of the present disclosure.

[0030] [Figure 4] An architecture for improving the discrimination of qubit states in a superconducting quantum computer by shifting the phase of a signal used to measure the state of a qubit, thereby removing the influence of noise when discriminating the qubit state, according to an embodiment of the present disclosure is shown.

[0031] [Figure 5] Calibration of a channel by a calibration engine according to an embodiment of the present disclosure is shown.

[0032] [Figure 6] A diagram showing a determination of whether the kernel response is affected by simultaneously performing quantum operations on qubits in adjacent channels in response to the completion of calibration of a new channel according to an embodiment of the present disclosure.

[0033] [Figure 7] This document shows one embodiment of the disclosure of a classic computer hardware configuration representing a hardware environment for implementing this disclosure.

[0034] [Figure 8] This is a flowchart of a method for calibrating a channel to train a kernel to have accurate calibration data, according to one embodiment of the present disclosure.

[0035] [Figure 9] This is a flowchart of a method for improving the identification of quantum states in a superconducting quantum computer according to one embodiment of the present disclosure. [Modes for carrying out the invention]

[0036] As described in the background technology section, in superconducting quantum computers, qubits such as transmon qubits are controlled via the use of microwave radio frequency signals. That is, such transmon qubits can be manipulated by microwave radio frequency signals. The state of such transmon qubits is then read, for example, via the use of a microwave resonator, where the resonant frequency of the resonator is shifted dispersively by the qubit state.

[0037] As the number of qubits increases, the useful frequency range becomes crowded and noisy, and effective computation diminishes. In particular, identifying transmon qubit states becomes difficult, especially when using dynamic circuits in kernel threshold systems.

[0038] A dynamic circuit is a quantum circuit that includes an intermediate circuit measurement, also known as a feedforward operation, which affects the control flow of subsequent gate executions in the circuit. The standard and simplest example is a conditional reset, where a projection measurement of a qubit is performed. If the result is |1>, the state is flipped from the |1> state to the |0> state using the X gate. If the result is |0>, the state is not flipped.

[0039] The dynamic circuit utilizes a low-latency decision architecture known as a "kernel," which performs evaluations based on the state read from the qubit. The "kernel" refers to a time-domain filter calibrated for a specific task. Next, real-time convolution of the arriving signal (e.g., a microwave radio frequency signal) is performed on the kernel to generate a combined result (also called the "kernel response") that is compared to a threshold to determine the quantum state of the qubit (e.g., a transmon qubit). For example, if the combined result exceeds the threshold, the result may be considered 1, and if the combined result is below the threshold, the result may be considered 0.

[0040] When calculations involving such dynamic circuits are extended to other states, such as |2> and |3> states, in addition to |0> and |1> states, it becomes difficult to maintain the low computational latency required to enable proper operation. As a result, a fast decision kernel is applied to sampled data in real time. Such outputs are compared to a set of thresholds to determine the state readout.

[0041] Unfortunately, kernel thresholding systems require a sufficiently large kernel sample size to clearly distinguish between input states. For example, the sample size can range from thousands to tens of thousands or even more. In latency-sensitive applications, such large sample sizes can result in long capture windows, thereby slowing down the responsiveness of dynamic circuits.

[0042] Furthermore, kernel threshold systems are often susceptible to in-band noise (e.g., crosstalk, Nyquist aliasing). Crosstalk is a disturbance caused by the electric or magnetic field of a signal affecting an adjacent signal. Nyquist aliasing refers to noise and signal images that are folded back from a higher Nyquist band to a first Nyquist band. As a result of such in-band noise, more samples are required, thereby further increasing the capture window and further slowing down the ability of the dynamic circuit to respond. Moreover, such noise can consequently make it difficult to identify qubit states (e.g., transmon qubit states).

[0043] In addition, in multiplexed applications, bleed-over energy (energy spread across adjacent channels) can introduce dynamic decision-making problems in a similar manner to crosstalk, including difficulties in identifying qubit states (e.g., transmonqubit states).

[0044] As a result, there is considerable interest in solutions to address the difficulty of identifying qubit states (e.g., transmonqubit states) in superconducting quantum computers, particularly those using dynamic circuits in kernel threshold systems.

[0045] Embodiments of the present disclosure provide means for improving the identification of qubit states in a superconducting quantum computer by shifting the phase of the signal used to measure the state of a qubit, thereby eliminating the effects of noise when identifying qubit states and increasing the speed and accuracy of state identification.

[0046] In some embodiments of this disclosure, the disclosure includes methods, systems, and computer program products for improving the identification of qubit states in a superconducting quantum computer. In one embodiment of this disclosure, a channel (qubit readout channel) is calibrated to train a kernel to include precise calibration data used with the readout signal to obtain a kernel response calculation. After calibrating the channel, a test is performed in which quantum operations are simultaneously performed on the qubits in adjacent channels, including the recently calibrated channel, to determine whether the kernel response differs from the expected kernel response. If such a situation occurs, the phase of the signal for the recently calibrated channel is shifted, and the process of recalibrating and testing the channel (using the phase-shifted signal) is repeated until the difference between the kernel response and the expected kernel response no longer exceeds a threshold. If such a situation occurs, the range of calculation for the kernel response is closest to the resonance point (the point where the signal frequency matches the natural frequency of the system), which effectively removes the effects of noise when identifying qubit states and improves the speed and accuracy of state identification.

[0047] The following description includes many specific details to provide a thorough understanding of the disclosure. However, it will be apparent to those skilled in the art that the disclosure can be put into practice even without such specific details. In other cases, well-known circuits are shown in block diagram form to avoid obscuring the disclosure with unnecessary details. In most cases, details considering timing considerations and similar considerations are omitted to the same extent because such details are not necessary for a complete understanding of the disclosure and are within the skill of those skilled in the art.

[0048] Referring here to the drawings in detail, Figure 1 shows one embodiment of the present disclosure of a communication system 100 for putting the principles of the present disclosure into practice. The communication system 100 comprises a quantum computer 101 configured to perform quantum computation, such as computation of the type that takes advantage of the collective properties of quantum states, such as superposition, interference and entanglement, and a classical computer 102 in which information is stored in bits that are logically represented by either 0 (off) or 1 (on). Examples of the classical computer 102 include, but are not limited to, portable computing units, personal digital assistants (PDAs), laptop computers, mobile devices, tablet personal computers, smartphones, mobile phones, navigation devices, gaming units, desktop computer systems, workstations, and the like, all configured to have the ability to connect to a network 113 (described later).

[0049] In one embodiment, a classical computer 102 is used to set up the qubit state in the quantum computer 101, after which the quantum computer 101 initiates the quantum process. Furthermore, in one embodiment, the classical computer 102 is configured to improve the identification of qubit states in the superconducting quantum computer by shifting the phase of the signal used to measure the qubit state, as will be further described later, thereby eliminating the influence of noise when identifying the qubit state.

[0050] In one embodiment, the hardware structure 103 of the quantum computer 101 includes a quantum data plane 104, a control and measurement plane 105, a control processor plane 106, a quantum controller 107, and a quantum processor 108.

[0051] The quantum data plane 104 includes physical qubits or quantum bits (the fundamental unit of quantum information, where a qubit is a two-state (or two-level) quantum mechanical system) and the structures necessary to hold them in place. In one embodiment, the quantum data plane 104 includes optional support circuits necessary for measuring the state of a qubit and performing gate operations on a physical qubit, or for controlling a Hamiltonian, for an analog computer. In one embodiment, a control signal routed to a selected qubit sets the state of the Hamiltonian. In the case of a gate-based system, since some qubit operations require two qubits, the quantum data plane 104 provides a programmable "wiring" network that allows two or more qubits to interact.

[0052] The control and measurement plane 105 converts the digital signals from the quantum controller 107, which indicate which quantum operations will be performed, into analog control signals necessary to perform operations on the qubits in the quantum data plane 104. In one embodiment, the control and measurement plane 105 converts the analog output of the qubit measurement in the quantum data plane 104 into classical binary data that can be handled by the quantum controller 107.

[0053] The control processor plane 106 identifies and triggers sequences of quantum gate operations and measurements (which are subsequently performed by the control and measurement plane 105 on the quantum data plane 104). These sequences execute programs provided by the quantum processor 108 for implementing quantum algorithms.

[0054] In one embodiment, the control processor plane 106 executes a quantum error correction algorithm (when the quantum computer 101 is performing error correction).

[0055] In one embodiment, the quantum processor 108 performs computational tasks using qubits. In certain domains where quantum mechanics operates, particles of matter can exist in multiple states, such as an "on" state, an "off" state, and both "on" and "off" states simultaneously. The quantum processor 108 leverages these quantum states of matter to output signals usable in data computing.

[0056] In one embodiment, the quantum processor 108 executes an algorithm that conventional processors cannot efficiently perform.

[0057] In one embodiment, the quantum processor 108 includes one or more quantum circuits 109. The quantum circuits 109 may be referred to collectively or individually as quantum circuits 109 or quantum circuit 109, respectively. As used herein, “quantum circuit 109” refers to a model for quantum computation in which the computation is a sequence of quantum logic gates, measurements, initialization of qubits to known values, and possibly other actions. In one embodiment, the quantum circuit 109 is a dynamic circuit. As used herein, a “dynamic circuit” is a quantum circuit that includes a circuit intermediate measurement, also known as a feedforward operation, which affects the control flow of subsequent gate executions within the circuit. The dynamic circuit utilizes a low-latency decision architecture known as a “kernel,” which makes evaluations based on the state read from the qubit. When computations relating to such a dynamic circuit are extended to other states, such as |2> and |3> states, in addition to the |0> and |1> states, it becomes difficult to maintain the low computational latency required to enable proper operation. As a result, a fast decision-making kernel is applied to the sampled data in real time. Such an output is compared to a set of thresholds to determine the state readout.

[0058] Furthermore, in one embodiment, the quantum circuit 109 corresponds to a command structure provided to the control processor plane 106, which specifies how to operate the control and measurement planes 105 to execute an algorithm on the quantum data plane 104 / quantum processor 108.

[0059] Furthermore, the quantum computer 101 includes a memory 110 which may correspond to a quantum memory. In one embodiment, the memory 110 is a set of qubits that store quantum states for later retrieval. The states stored in the quantum memory 110 can hold quantum superpositions.

[0060] In one embodiment, memory 110 stores an application 111 which may be configured to carry out one or more of the methods described herein in one or more embodiments. For example, application 111 may implement a program to improve the identification of qubit states in a superconducting quantum computer by shifting the phase of the signal used to measure the state of a qubit, thereby eliminating the influence of noise when identifying the qubit state, as will be further described in relation to Figures 2 to 9. Examples of memory 110 include optical quantum memory, solid-state quantum memory, gradient echo memory, and electromagnetic induction transparency.

[0061] Furthermore, in one embodiment, the classical computer 102 includes a “transpiler 112,” which, when used herein, is configured to rewrite the abstract quantum circuit 109 into a functionally equivalent that matches the constraints and characteristics of a particular target quantum device. In one embodiment, the transpiler 112 (e.g., qiskit.transpiler, where Qiskit® is an open-source software development kit for working with quantum computers at the circuit, pulse, and algorithm level) translates a machine learning model trained at runtime on quantum hardware 103 into its basic instructions and maps them to physical qubits.

[0062] In one embodiment, the quantum machine learning model is based on a variational quantum circuit 109. Such a model consists of data encoding, processing parameterized with trainable parameters, and measurement / post-processing.

[0063] In one embodiment, the number of qubits (the fundamental unit of quantum information, where a qubit is a two-state (or two-level) quantum mechanical system) is determined by the number of features in the data. This processing step may include multiple layers of parameterized gates. As a result, in one embodiment, the number of trainable parameters is (number of features) * (number of layers).

[0064] Furthermore, as shown in Figure 1, a classical computer 102 used to set up the state of qubits in the quantum computer 101 can be connected to the quantum computer 101 via a network 113.

[0065] Network 113 could be, for example, a quantum network, a local area network, a wide area network, a wireless wide area network, a circuit-switched telephone network, a Global System for Mobile Communications (GSM) network, a Wireless Application Protocol (WAP) network, a WiFi network, an IEEE 802.11 standard network, a cellular network, and various combinations thereof. Other networks, whose descriptions are omitted herein for the sake of brevity, may also be used in conjunction with System 100 in Figure 1 without departing from the scope of this disclosure.

[0066] Furthermore, the classical computer 102 is configured to improve the identification of qubit states in the superconducting quantum computer by shifting the phase of the signal used to measure the qubit state, thereby eliminating the influence of noise when identifying the qubit state, as will be further described in relation to Figures 2 to 9. A description of the software components of the classical computer 102 is provided below in relation to Figure 3, and a description of the hardware configuration of the classical computer 102 is provided below in relation to Figure 7.

[0067] System 100 is not limited to any one specific network architecture. System 100 may include any number of quantum computers 101, classical computers 102, and networks 113.

[0068] A brief discussion regarding the influence on thresholds is considered appropriate here.

[0069] As mentioned earlier, the fast decision-making kernel for dynamic circuits is applied in real time to sampled data. Such an output (also called the "integrated result" or "kernel response") is compared against a set of thresholds to determine the state readout. Unfortunately, kernel thresholding systems require a sufficiently large kernel sample size to clearly distinguish between input states. Furthermore, such kernel thresholding systems are often susceptible to in-band noise such as crosstalk or multiplexing.

[0070] It has been found that the kernel response from a signal (e.g., a microwave radio frequency (RF) signal) has a larger threshold as it approaches the resonance point (the point where the signal frequency matches the system's natural frequency). Exemplary kernel response diagrams based on frequency and phase range from 0 to 2*π are provided in Figure 2.

[0071] Referring to Figure 2, Figure 2 shows the kernel response 200 in terms of the range of calculations ("kernel response calculation" or "integrated result calculation") shown along the Y-axis 201 (indicated by the length of vertical bars such as bar 202) for the readout threshold, with respect to the frequency and phase of the signal (microwave RF signal used to measure qubits) shown along the X-axis 203, and respect to the readout threshold.

[0072] As shown in Figure 2, the range of kernel response calculations for the readout threshold, such as that indicated by bar 202, increases as it approaches the resonance point (the point where the signal frequency matches the system's natural frequency).

[0073] Furthermore, the effect on the threshold is altered by shifting the phase of the victim or aggressor. As used herein, “aggressor” refers to the noise source; “victim” refers to the signal affected by the aggressor's noise. By phase-shifting the aggressor or victim signal, the kernel response 200 can yield a wider range of calculations (kernel response calculation or integrated result calculation) for calculating the readout threshold, particularly near the resonance point.

[0074] As a result, the principle of this disclosure addresses in-band noise such as crosstalk or multiplexing by shifting the phase of the signal used to measure the qubit so that the kernel response approaches the resonance point, as will be discussed later in relation to Figure 3, thereby eliminating the effect of noise when identifying the qubit state.

[0075] Figure 3 shows a software component of a classical computer 102 (Figure 1) used in one embodiment of the present disclosure to improve the identification of qubit states in a superconducting quantum computer by shifting the phase of the signal used to measure the qubit state, thereby eliminating the influence of noise when identifying the qubit state.

[0076] Referring to Figure 3 in conjunction with Figure 1, the classical computer 102 includes a calibration engine 301 configured to calibrate a channel ("qubit readout channel") used to communicate the measured state of a qubit. As used herein, "channel" refers to a radio frequency channel used to communicate the measured state of a qubit. An example of such a channel will be described later in relation to Figures 4 and 5.

[0077] Figure 4 shows an architecture according to one embodiment of the present disclosure for improving the identification of qubit states in a superconducting quantum computer by shifting the phase of the signal used to measure the state of a qubit, thereby eliminating the influence of noise when identifying the qubit state. As shown in Figure 4, such an architecture includes channels 401A to 401N connected to a state test matrix 402, where N is a positive integer. Channels 401A to 401N may be referred to collectively or individually as channels 401 or channel 401, respectively.

[0078] As described above, channel 401 is used to read the state of the qubit. Furthermore, in one embodiment, channel 401 is calibrated to train a kernel to have accurate calibration data used with the read signal to obtain a kernel response calculation that determines the qubit state when compared to a threshold. In one embodiment, the calibration of channel 401 includes setting the excited state of the qubit to a specific state (e.g., a qubit state of 1), and then reading the qubit's response for that state a user-specified number of iterations used to calibrate the kernel for that specific state. Such a process is performed for each qubit state. Such kernel response calculations are used, with a state test matrix 402, as will be discussed later, to determine whether channel 401 is affected by in-band noise. Further discussion of the calibrated channel 401 is provided below in relation to Figure 5.

[0079] In one embodiment, the state test matrix 402 is configured to test whether the kernel response calculation is affected by performing quantum operations on qubits simultaneously in adjacent channels 401, such as a recently calibrated channel 401 (e.g., channel 401A) and a previously calibrated channel 401 (e.g., channel 401B), using a kernel trained for the qubit states for each calibrated channel 401. Since the trained kernel contains accurate kernel state calculations, an accurate kernel response calculation is known that results in an accurate determination of the quantum state of the qubit. Such a kernel response calculation can then be compared to the kernel response calculation obtained from the state test matrix 402. If the difference between such kernel response calculations exceeds a threshold that may be user-specified, such a difference can be inferred to be due to in-band noise. In an attempt to eliminate the effects of such noise while identifying qubit states and improving the speed and accuracy of state identification, the phase of the signal in the calibrated channel 401 (e.g., channel 401A) is shifted in response to the range of the readout threshold calculation (i.e., kernel response calculation) not meeting such criteria. As used herein, “criterion” refers to the difference between kernel response calculations being smaller than a user-defined threshold. After shifting the phase of the signal for a recently calibrated channel 401, the process of calibrating such channel 401 using the phase-shifted signal is repeated to train a kernel, and then the trained kernel is used to determine (“test”) whether further phase adjustment of the phase-shifted signal is needed by the state test matrix 402. A further discussion of the test by the state test matrix 402 is provided below in relation to Figure 6.

[0080] Referring now to Figure 5, Figure 5 shows the calibration of channel 401 by the calibration engine 301 according to one embodiment of the present disclosure. In particular, Figure 5 shows the procedure flow for calibrating the kernel based on channel 401. As shown in Figure 5, the calibration engine 301 sets the excited state of a qubit (e.g., a transmon qubit) to a specific state (higher energy than the ground state) by applying a signal (e.g., an RF signal) to the qubit (see element 501). An excited state is any quantum state (e.g., |1>, |2>) that has a higher energy than the ground state, |0>. In one embodiment, the setting points for each state are known from the previous calibration.

[0081] The calibration engine 301 is further configured to read out the response of a qubit as a result of a dynamic circuit performing a quantum operation on the qubit (see element 502). That is, the calibration engine 301 reads out the response of a qubit set to a particular qubit state (e.g., the qubit state of |0>). In one embodiment, such a measurement is compiled (see element 503) and stored in a data structure or the like stored in the storage device of the classical computer 102. As will be described later, such a measurement is used to obtain a kernel state calculation of the kernel for the particular qubit state of interest.

[0082] Furthermore, in one embodiment, the calibration engine 301 resets the qubit by setting it to the ground state in preparation for setting its excited state in the next cycle (see element 504).

[0083] Furthermore, in one embodiment, after each readout of the qubit's response, the calibration engine 301 sets the qubit's excited state to a specific state (e.g., state |1>) and then determines whether a threshold number of readouts (a threshold number of cycles) has been performed (see element 505). In one embodiment, such a threshold number of cycles is user-specified.

[0084] If the threshold number of cycles has not been reached, another iteration is performed as described above, in which the response of the qubit is read out as a result of the dynamic circuit performing a quantum operation on the qubit set to the same quantum state (e.g., the quantum state of |1>). That is, the qubit is excited to the same excited state as the previous excitation. This process is repeated at least N times, where N is a positive integer that may be user-specified. In other words, the process of exciting the qubit, reading or measuring the qubit's response, and resetting the qubit is performed N times for a particular qubit state.

[0085] However, if the threshold number of cycles has been reached, the calibration engine 301 determines whether there are any new states (quantum states) that have not been used to train the kernel (see element 506). That is, the calibration engine 301 determines whether the kernel has been trained to obtain kernel state calculations for such quantum states (e.g., quantum states of |0>).

[0086] If there are new states that will be used to train the kernel, the calibration engine 301 sets the new state values ​​as excited states of the qubits (see elements 507 and 501), and the process involving elements 501-504 is repeated as described above.

[0087] However, if there are no new states to be used to train the kernel, the calibration engine 301, having completed the calibration of channel 401, generates a "kernel state calculation" for each qubit state (e.g., the qubit state of |0>, the qubit state of |1>) that will be stored in the kernel (see element 508).

[0088] In one embodiment, such a kernel state calculation corresponds to a read response of a qubit set to a specific state (e.g., the qubit state of |0>) as described above (see element 502). Thus, the calibration engine 301 generates a kernel state calculation based on such read responses that will be stored in the kernel for each qubit state (e.g., the qubit state of |0>).

[0089] As mentioned above, the kernel response calculation ("integrated response calculation") is compared to a threshold to determine the quantum state of the qubit. Since the qubit state is known when calibrating channel 401, such a kernel state calculation can be derived using the procedure described above. Therefore, the kernel response calculation ("integrated response calculation") that should be obtained to accurately measure the state of the qubit is also known based on the kernel training described above.

[0090] In one embodiment, the classical computer 102 includes a test engine 302 configured to use these kernel state calculations of the kernel to determine whether in-band noise is affecting the identification of quantum states by performing quantum operations on qubits simultaneously in adjacent channels 401 in response to the completion of calibration of a new channel 401, as described later in relation to Figure 6.

[0091] Figure 6 shows a determination of whether the kernel response is affected by simultaneously performing quantum operations on qubits in adjacent channels 401 in response to the completion of calibration of a new channel 401, according to one embodiment of the present disclosure.

[0092] As shown in Figure 6, the test engine 302 determines whether the new quantum channel 401 has completed calibration (see element 601). If the new quantum channel 401 has not completed calibration, the test engine 302 continues to determine whether the new quantum channel 401 has completed calibration. However, if the new quantum channel 401 has completed calibration, the test is performed as described below.

[0093] In one embodiment, the test engine 302 reads the response of a qubit set to a specific qubit state (e.g., the |0> qubit state) for a recently calibrated channel 401 (e.g., channel 401A), and simultaneously reads the response of a qubit set to a specific qubit state (e.g., the |1> qubit state) for an adjacent channel 401 (e.g., channel 401B). In this way, the principle of the present disclosure determines whether in-band noise affecting quantum state identification is affecting the kernel response calculation, as will be described later.

[0094] In one embodiment, the test engine 302 sets the state of qubits in adjacent calibrated channels (e.g., channels 401A, 401B), including one of the channels that has recently completed calibration (e.g., channel 401A). For example, for qubits "Q0" and "Q1" having two states, such qubits may be set to the following state values: |00>, |01>, |10>, |11>, where qubit Q0 may be set to the value |0> for the recently calibrated channel 401 (e.g., channel 401A), and qubit Q1 may be set to the value |1> for the previously calibrated channel 401 (e.g., channel 401B). Next, the test engine 302 may perform readouts of responses to such qubits for such channels 401, as shown in units 602A to 602B, where unit 602A corresponds to performing readouts of responses to qubits for channels 401 that have recently been calibrated (e.g., channel 401A), and unit 602B corresponds to performing readouts of responses to qubits for channels 401 that have previously been calibrated (e.g., channel 401B). Units 602A to 602B may be referred to as unit 602 or unit 602, respectively, collectively or individually.

[0095] Referring to unit 602 in Figure 6, the test engine 302 sets the excited state of a qubit (e.g., a transmon qubit) to a specific state (e.g., a qubit state of |0>) by applying a signal (e.g., an RF signal) to the qubit (see element 603).

[0096] The test engine 302 is further configured to read out the qubit's response as a result of a dynamic circuit performing a quantum operation on the qubit (see element 604). That is, the test engine 302 reads out the qubit's response set to a specific qubit state in each of the channels 401. In one embodiment, such measurements are compiled (see element 605) and stored in a data structure or the like stored in the storage device of the classical computer 102.

[0097] Furthermore, in one embodiment, the test engine 302 resets the qubit by setting it to the ground state in preparation for setting its excited state in the next test (see element 606).

[0098] In one embodiment, the test engine 302 determines a kernel response calculation involving a recently calibrated channel 401 (e.g., channel 401A). Such a kernel response calculation is then compared to a kernel response calculation obtained from the aforementioned calibrated channel 401 (e.g., channel 401A) (see element 607). Such a comparison is performed to determine whether there is a threshold effect (see element 608). That is, such a comparison is performed to determine whether the difference between such kernel response calculations exceeds a threshold. In other words, the comparison is performed to determine whether the kernel response meets or fails to meet a user-defined criterion. As used herein, “criterion” means that the difference between kernel response calculations is less than a user-defined threshold.

[0099] If the difference exceeds a threshold (i.e., there is a threshold effect), the phase of the signal in the most recently calibrated channel 401 (e.g., channel 401A) is shifted (see element 609). The calibrated channel 401 (e.g., channel 401A) is then recalibrated using the phase-shifted signal as described above.

[0100] However, if the difference does not exceed the threshold (i.e., there is no threshold effect), the phase of the signal in the recently calibrated channel 401 (e.g., channel 401A) is neither shifted nor adjusted (i.e., the phase of the signal previously shifted in the previous iteration is not adjusted) (see element 610). The signal, which then includes the previously phase-shifted signal, can then be used to measure the state of the qubit.

[0101] The process described above continues until the signal phase shift achieves the desired result (i.e., the difference between kernel response calculations does not exceed a threshold). If such a desired result cannot be achieved, an error is reported, for example, to the user of the classical computer 102.

[0102] Further discussion of these and other features is provided below in connection with the discussion of methods for improving the identification of qubit states.

[0103] Before discussing methods for improving the identification of qubit states, a description of the hardware configuration of the classical computer 102 (Figure 1) is provided below in relation to Figure 7.

[0104] Referring here to Figure 7 in conjunction with Figure 1, Figure 7 shows one embodiment of the present disclosure, which represents a hardware configuration of a classic computer 102 representing a hardware environment for carrying out the present disclosure.

[0105] Various aspects of this disclosure are illustrated by narrative text, flowcharts, block diagrams of computer systems, and / or block diagrams of mechanical logic included in embodiments of computer program products (CPPs). With respect to any flowchart, depending on the technology involved, operations may be performed in a different order than those shown in a given flowchart. For example, again depending on the technology involved, two operations shown in consecutive blocks of a flowchart may be performed in reverse order, as a single integrated step, simultaneously, or with at least partial time overlap.

[0106] Embodiments of a computer program product ("CPP Embodiment" or "CPP") are terms used in this disclosure and describe any set of one or more storage media (also called "Multiple Media") that are collectively comprised of a set of one or more storage devices that collectively comprise machine-readable code corresponding to instructions and / or data for performing computer operations specified in a given CPP claim. A "Storage Device" is any tangible device capable of holding and storing instructions for use by a computer processor. Computer-readable storage media may be, but are not limited to, electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, mechanical storage media, or any preferred combination thereof. Some known types of storage devices, including these media, include diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), compact disk read-only memory (CD-ROM), digital versatile disks (DVDs), memory sticks, floppy disks, mechanically encoded devices (e.g., pits / lands formed on the main surface of punch cards or disks), or any preferred combination of those described above. Computer-readable storage media, as used in this disclosure, shall not be construed as storage in the form of transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides, optical pulses passing through optical fiber cables, electrical signals communicated through wires, and / or other transmission media. As will be understood by those skilled in the art, data is moved at several intermittent points in the normal operation of a storage device, such as during access, defragmentation, or garbage collection; however, data is not transient while it is stored, so the foregoing does not make a storage device transient.

[0107] The computing environment 700 includes an example of an environment for the execution of at least some computer code (stored in block 701) involved in the execution of inventive methods, such as improved identification of qubit states. In addition to block 701, the computing environment 700 includes, for example, a classical computer 102, a network 113 such as a wide area network (WAN), an end-user device (EUD) 702, a remote server 703, a public cloud 704, and a private cloud 705. In this embodiment, the classical computer 102 includes a processor set 706 (including processing circuits 707 and a cache 708), a communication fabric 709, volatile memory 710, persistent storage 711 (including an operating system 712 and block 701 as identified above), a peripheral device set 713 (including a user interface (UI) device set 714, storage 715, and an Internet of Things (IoT) sensor set 716), and a network module 717. The remote server 703 includes a remote database 718. Public Cloud 704 includes a gateway 719, a cloud orchestration module 720, a host physical machine set 721, a virtual machine set 722, and a container set 723.

[0108] The classical computer 102 may take the form of a desktop computer, laptop computer, tablet computer, smartphone, smartwatch or other wearable computer, mainframe computer, quantum computer, or any other form of computer or mobile device currently known or to be developed in the future that is capable of running programs, accessing networks, or querying databases such as the remote database 718. As is well understood in the field of computer technology, and depending on the technology, the execution of a computer implementation method may be distributed among multiple computers and / or multiple locations. On the other hand, in this presentation of the computing environment 700, in order to keep the presentation as simple as possible, the detailed discussion focuses on a single computer, specifically the classical computer 102. The classical computer 102 may be located in the cloud, even if it is not shown in the cloud in Figure 7. On the other hand, the classical computer 102 does not need to be in the cloud, except to any extent that may be definitively shown.

[0109] The processor set 706 includes one or more computer processors of any type currently known or to be developed in the future. The processing circuitry 707 may be distributed across multiple packages, for example, multiple interconnected integrated circuit chips. The processing circuitry 707 may implement multiple processor threads and / or multiple processor cores. The cache 708 is memory located within the processor chip package and is typically used for data or code that should be available for high-speed access by threads or cores running on the processor set 706. The cache memory is typically organized into multiple levels depending on its relative proximity to the processing circuitry. Alternatively, some or all of the cache for the processor set may be located "off-chip". In some computing environments, the processor set 706 may operate using qubits and be designed to perform quantum computing.

[0110] Computer-readable program instructions are typically loaded into a classical computer 102, causing the processor set 706 of the classical computer 102 to execute a series of operational steps, thereby effectively implementing the computer implementation method, and as a result, the instructions thus executed instantiate the method specified in the flowcharts and / or descriptive descriptions of the computer implementation method contained herein (collectively referred to as the “inventive method”). These computer-readable program instructions are stored in various types of computer-readable storage media, such as the cache 708 and other storage media described later. The program instructions and associated data are accessed by the processor set 706 to control and direct the execution of the inventive method. In the computing environment 700, at least some of the instructions for executing the inventive method may be stored in block 701 in persistent storage 711.

[0111] The communication fabric 709 is a signal conduction path that enables various components of a classic computer 102 to communicate with one another. Typically, this fabric is made up of switches and conductive paths, such as buses, bridges, and physical input / output ports. Other types of signal communication paths, such as fiber optic communication paths and / or wireless communication paths, may be used.

[0112] The volatile memory 710 is any type of volatile memory currently known or to be developed in the future. Examples include dynamic type random access memory (RAM) or static type RAM. Typically, volatile memory is characterized by random access, but this is not mandatory unless explicitly stated. In a classical computer 102, the volatile memory 710 is located in a single package and is internal to the classical computer 102, but alternatively or in addition, the volatile memory may be distributed across multiple packages and / or located externally to the classical computer 102.

[0113] Persistent storage 711 is any form of non-volatile storage for a computer, currently known or to be developed in the future. Non-volatility of this storage means that the stored data is maintained whether or not power is supplied to the classic computer 102 and / or directly to the persistent storage 711. Persistent storage 711 may be read-only memory (ROM), but typically at least a portion of the persistent storage allows for writing, deleting, and rewriting of data. Some well-known forms of persistent storage include magnetic disks and solid-state storage devices. The operating system 712 may take several forms, such as various known proprietary operating systems or open-source portable operating system interface type operating systems that utilize a kernel. The code contained in block 701 typically includes at least some computer code involved in the execution of an inventive method.

[0114] The peripheral device set 713 includes a set of peripheral devices for the classical computer 102. Data communication connections between the peripheral devices and other components of the classical computer 102 can be implemented in various ways, such as Bluetooth connections, near-field communication (NFC) connections, cable connections (such as Universal Serial Bus (USB) type cables), insert-type connections (e.g., Secure Digital (SD) cards), connections made through local area communication networks, and even connections made through wide area networks such as the Internet. In various embodiments, the UI device set 714 may include components such as a display screen, speakers, microphones, wearable devices (such as goggles and smartwatches), keyboards, mice, printers, touchpads, game controllers, and haptic devices. Storage 715 is external storage such as an external hard drive, or insertable storage such as an SD card. Storage 715 may be persistent and / or volatile. In some embodiments, storage 715 may take the form of a quantum computing storage device for storing data in the form of qubits. In embodiments where the classical computer 102 requires a large amount of storage (for example, when the classical computer 102 locally stores and manages a large database), this storage may be provided by peripheral storage devices designed to store very large amounts of data, such as a Storage Area Network (SAN) shared by multiple geographically distributed computers. The IoT sensor set 716 consists of sensors that can be used in Internet of Things applications. For example, one sensor may be a thermometer and another may be a motion detector.

[0115] The network module 717 is a collection of computer software, hardware, and firmware that enables the classical computer 102 to communicate with other computers through the WAN 113. The network module 717 may include hardware such as a modem or Wi-Fi signal transceiver, software for packetizing and / or depacketizing data for communication network transmission, and / or web browser software for communicating data over the internet. In some embodiments, the network control and network forwarding functions of the network module 717 are performed on the same physical hardware device. In other embodiments (e.g., embodiments utilizing software-defined networking (SDN)), the control and forwarding functions of the network module 717 are performed on physically separate devices, such that the control function manages several different network hardware devices. Computer-readable program instructions for performing the inventive method can typically be downloaded from an external computer or external storage device to the classical computer 102 through a network adapter card or network interface included in the network module 717.

[0116] WAN113 is any wide area network (e.g., the Internet) that can transmit computer data over non-local distances using any currently known or future-developed technology for transmitting computer data. In some embodiments, a WAN may be replaced and / or supplemented by a local area network (LAN), such as a Wi-Fi network, designed to transmit data between devices located in a local area. WANs and / or LANs typically include computer hardware such as copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and edge servers.

[0117] The end-user device (EUD) 702 is any computer system used and controlled by an end-user (e.g., a customer of the company operating the classical computer 102) and can take any of the forms described above in relation to the classical computer 102. Typically, the EUD 702 receives useful data from the operation of the classical computer 102. For example, in a hypothetical case where the classical computer 102 is designed to provide recommendations to the end-user, these recommendations are typically communicated to the EUD 702 from the network module 717 of the classical computer 102 via the WAN 113. In this way, the EUD 702 can display or otherwise present the recommendations to the end-user. In some embodiments, the EUD 702 may be a client device such as a thin client, a heavy client, a mainframe computer, and a desktop computer.

[0118] The remote server 703 is any computer system that provides at least some data and / or functionality as a service to the classical computer 102. The remote server 703 may be controlled and used by the same entity that operates the classical computer 102. The remote server 703 represents a machine that collects and stores useful data for use by other computers, such as the classical computer 102. For example, in a hypothetical case where the classical computer 102 is designed and programmed to provide recommendations based on historical data, this historical data may then be provided to the classical computer 102 from the remote database 718 of the remote server 703.

[0119] Public Cloud 704 is any computer system available for use by multiple entities, providing on-demand availability of computer system resources and / or other computer functions, particularly data storage (cloud storage) and computing power, without direct active management by the user. Cloud computing typically leverages resource sharing to achieve coherence and economies of scale. Direct and active management of Public Cloud 704's computing resources is performed by the computer hardware and / or software of Cloud Orchestration Module 720. The computing resources provided by Public Cloud 704 are typically implemented by virtual computing environments running on various computers that make up the host physical machine set 721, which is a universe of physical computers residing within and / or available to Public Cloud 704. Virtual computing environments (VCEs) typically take the form of virtual machines from the virtual machine set 722 and / or containers from the container set 723. These VCEs may be stored as images and can be transferred either as images or after instantiation of the VCEs, among and between hosts of various physical machines. The cloud orchestration module 820 manages image transfer and storage, deploys new VCE instances, and manages the active instantiation of VCE deployments. The gateway 719 is a collection of computer software, hardware, and firmware that enables the public cloud 704 to communicate over the WAN 113.

[0120] Here, some further explanation of virtualized computing environments (VCEs) is provided. A VCE can be stored as an "image." A new active instance of a VCE can be instantiated from an image. Two well-known types of VCEs are virtual machines and containers. A container is a VCE that uses operating system-level virtualization. This refers to an operating system feature where the kernel allows for the existence of multiple isolated user-space instances called containers. These isolated user-space instances typically behave like actual computers in terms of the programs running within them. Computer programs running on a normal operating system can utilize all the resources of that computer, including connected devices, files and folders, network shares, CPU power, and quantifiable hardware capabilities. However, programs running inside a container can only use the contents of the container and the devices allocated to the container; this feature is known as containerization.

[0121] Private Cloud 705 is similar to Public Cloud 704, except that its computing resources are available only for use by a single enterprise. While Private Cloud 705 is described in other embodiments as communicating with WAN 113, a private cloud may be completely isolated from the internet and accessible only through a local / private network. A hybrid cloud is a combination of multiple clouds of different types (e.g., private, community, or public cloud types), often implemented by different vendors. Each of the multiple clouds remains a separate, discrete entity, but the larger hybrid cloud architecture is bound together by standardized or proprietary technologies that enable orchestration, management, and / or data / application portability between the multiple configured clouds. In this embodiment, both Public Cloud 704 and Private Cloud 705 are part of a larger hybrid cloud.

[0122] Block 701 further includes the software components described above in relation to Figures 3 to 6 in order to improve the identification of qubit states. In one embodiment, such components may be implemented in hardware. The functions described above, performed by such components, are not general-purpose computer functions. As a result, the classic computer 102 is a specific machine that is the result of the implementation of specific, non-general-purpose computer functions.

[0123] In one embodiment, the functionality of such a software component of a classical computer 102, including functionality for improving the identification of qubit states, can be embodied in an application-specific integrated circuit.

[0124] As mentioned above, in superconducting quantum computers, qubits such as transmon qubits are controlled via the use of microwave radio frequency signals. That is, such transmon qubits can be manipulated by microwave radio frequency signals. The state of such transmon qubits is then read, for example, via the use of a microwave resonator, where the resonant frequency of the resonator is shifted dispersively by the qubit state. As the number of qubits increases, the useful frequency range becomes crowded and noisy, and effective computation diminishes. In particular, it becomes difficult to identify the transmon qubit state using dynamic circuits, especially in kernel threshold systems. Dynamic circuits are quantum circuits that include circuit intermediate measurements, also known as feedforward operations, which affect the control flow of gate execution later in the circuit. A standard and simplest example is a conditional reset, where a projection measurement of the qubit is performed. If the result is |1>, the state is flipped from the |1> state to the |0> state using an X gate. If the result is |0>, the state is not flipped. Dynamic circuits utilize a low-latency decision architecture known as a "kernel," which performs evaluations based on the state read from a qubit. The "kernel" refers to a time-domain filter calibrated for a specific task. Real-time convolution of the incoming signal (e.g., a microwave radio frequency signal) is then performed on the kernel to generate a combined result (also called the "kernel response") which is compared to a threshold to determine the quantum state of the qubit (e.g., a transmon qubit). For example, if the combined result exceeds the threshold, the result may be considered |1>, and if the combined result is below the threshold, the result may be considered |0>. When computations for such dynamic circuits are extended to other states, such as |2> and |3>, in addition to |0> and |1> states, it becomes difficult to maintain the low computational latency required to enable proper operation. As a result, a fast decision kernel is applied to sampled data in real time. Such an output is compared to a set of thresholds to determine the state readout. Unfortunately, kernel thresholding systems require a sufficiently large kernel sample size to clearly distinguish the input states. For example, sample sizes can range from thousands to tens of thousands or even more. In latency-sensitive applications, such large sample sizes can result in long capture windows, thereby slowing down the responsiveness of dynamic circuits. Furthermore, kernel threshold systems are often susceptible to in-band noise (e.g., crosstalk, Nyquist aliasing). Crosstalk is a disturbance caused by the electric or magnetic field of a signal affecting adjacent signals. Nyquist aliasing refers to noise and signal images that are folded back from the upper Nyquist band to the first Nyquist band. As a result of such in-band noise, more samples are required, thereby further increasing the capture window and further slowing down the responsiveness of dynamic circuits. Moreover, such noise can consequently make it difficult to identify qubit states (e.g., transmon qubit states).In addition, in multiplexed applications, bleed-over energy (energy spread across adjacent channels) can introduce dynamic decision-making problems in a similar manner to crosstalk, including difficulties in identifying qubit states (e.g., transmonqubit states).

[0125] Embodiments of the present disclosure provide means for identifying qubit states in a superconducting quantum computer by using dynamic circuits in a kernel threshold system, in particular, to shift the phase of a signal in response to a kernel response failing to meet a criterion, as described later in relation to Figures 8-9. Figure 8 is a flowchart of a method for calibrating a channel to train a kernel to have accurate calibration data. Figure 9 is a flowchart of a method for improving the identification of quantum states in a superconducting quantum computer.

[0126] As described above, Figure 8 is a flowchart of a method 800 for calibrating channel 401 (Figures 4 and 5) to train a kernel to have accurate calibration data, according to one embodiment of the present disclosure.

[0127] Referring to Figure 8, and in conjunction with Figures 1-7, in step 801, the calibration engine 301 sets the excited state of a qubit (e.g., a transmon qubit) to a specific quantum state (e.g., a |0> quantum state) by applying a signal to the qubit for the channel 401 to be calibrated (e.g., channel 401A).

[0128] As described above, the calibration engine 301 sets the excited state of a qubit to a specific state (with higher energy than the ground state) (e.g., the quantum state of |0>) by applying a signal (e.g., an RF signal) to the qubit. An excited state is any quantum state (e.g., |1>, |2>) with higher energy than the ground state, |0>. In one embodiment, the setting points for each state are determined from the previous calibration.

[0129] In step 802, the calibration engine 301 reads out the response of the qubit, which has been set to a specific state (e.g., the quantum state of |0>) as a result of the dynamic circuit performing a quantum operation on the qubit.

[0130] In step 803, the calibration engine 301 compiles the results of the readout (measured value).

[0131] As mentioned above, such measurements are compiled and stored in data structures on the storage devices of the classical computer 102 (e.g., storage devices 711, 715). As will be discussed later, such measurements are used to obtain kernel state calculations for the kernel for a particular qubit state of interest.

[0132] In step 804, the calibration engine 301 resets the qubit by setting it to the ground state in preparation for setting its excited state in the next cycle.

[0133] In step 805, the calibration engine 301 determines whether it has reached a user-specified threshold number of cycles for reading out the response of a qubit set to a specific state (e.g., the quantum state of |0>). For example, in one embodiment, the calibration engine 301 reads out the response of a qubit set to a specific state (e.g., the quantum state of |0>) for N cycles, where N is a positive integer. As will be described later, the calibration engine 301 performs such readouts of qubits for each specific quantum state.

[0134] If the threshold number of cycles has not been reached, the calibration engine 301 sets the excited state of the qubit for channel 401 being calibrated (e.g., channel 401A) to the same quantum state (e.g., the |0> quantum state) in step 801, and then reads out the qubit's response in step 802. The process in steps 801-804 continues until the threshold number of cycles is reached.

[0135] If the threshold number of cycles is reached, in step 806, the calibration engine 301 determines whether there are any new states (quantum states) that have not been used to train the kernel. That is, the calibration engine 301 determines whether the kernel has been trained to obtain kernel state calculations for such quantum states (e.g., the quantum state of |1>).

[0136] If there is a new quantum state to be used to train the kernel, in step 807 the calibration engine 301 sets the new quantum state value (e.g., the quantum state of |1>) as the excited state. Then the calibration engine 301 sets the excited state of the qubit (e.g., the quantum state of |1>) as the new quantum state value, for example, by applying a signal (e.g., an RF signal) to the qubit in step 801. Next, the process in steps 801-804 is repeated for this quantum state (e.g., the quantum state of |1>) until a threshold number of cycles is reached.

[0137] However, if there are no new states to be used to train the kernel, in step 808, the calibration engine 301 generates kernel states that will be stored in the kernel for each qubit state (e.g., qubit state of |0>, qubit state of |1>) now that the calibration of channel 401 is complete. In one embodiment, such kernel state calculations are recorded in the storage devices of the classical computer 102 (e.g., storage devices 711, 715).

[0138] As described above, in one embodiment, such a kernel state calculation corresponds to a readout response of a qubit in response to setting the qubit to a specific state (e.g., the qubit state of |0>), as previously stated. Thus, the calibration engine 301 generates a kernel state calculation to be stored in the kernel for each qubit state (e.g., the qubit state of |0>) based on such readout responses.

[0139] As mentioned above, the kernel response calculation ("integrated response calculation") is compared to a threshold to determine the quantum state of the qubit. Since the qubit state is known when calibrating channel 401, such a kernel state calculation can be derived using the procedure described above. Therefore, the kernel response calculation ("integrated response calculation") that should be obtained to accurately measure the state of the qubit is also known based on the kernel training described above.

[0140] As will be discussed later, in conjunction with Figure 9, such a trained kernel may be used to determine whether it is necessary to shift the phase of the signal in calibrated channel 401 (e.g., channel 401A) to eliminate the adverse effects of in-band noise and bleed-over energy, thereby improving the identification of qubit states (e.g., transmon qubit states) in a superconducting quantum computer.

[0141] Figure 9 is a flowchart of a method 900 for improving the identification of quantum states in a superconducting quantum computer according to one embodiment of the present disclosure.

[0142] Referring to Figures 1-8 and Figure 9, in step 901, the test engine 302 sets the excited state of a qubit (e.g., a transmon qubit) in adjacent calibrated channels (e.g., channels 401A, 401B), including the recently calibrated channel 401 (e.g., channel 401A), to a specific quantum state, as shown in units 602A-602B of Figure 6. In one embodiment, the test engine 302 sets the excited state of a qubit to a specific state (e.g., a qubit state of |0>) by applying a signal (e.g., an RF signal) to the qubit.

[0143] For example, for two qubits "Q0" and "Q1" having two states, such qubits may be set to the following state values: |00>, |01>, |10>, and |11>, where qubit Q0 may be set to the value |0> for the most recently calibrated channel 401 (e.g., channel 401A), and qubit Q1 may be set to the value |1> for the previously calibrated channel 401 (e.g., channel 401B).

[0144] In step 902, the test engine 302 reads out the responses to such qubits for such channels 401 as a result of the dynamic circuit performing quantum operations on the qubits, as shown in units 602A to 602B of Figure 6, where unit 602A corresponds to reading out the qubit response for a channel 401 that has recently completed calibration (e.g., channel 401A), and where unit 602B corresponds to reading out the qubit response for a channel 401 that has previously completed calibration (e.g., channel 401B). That is, the test engine 302 reads out the qubit responses set to a specific qubit state for each of the channels 401 being tested. In one embodiment, such measurements are compiled and stored in a data structure or the like, stored in a storage device of the classical computer 102 (e.g., storage devices 711, 715).

[0145] In step 903, the test engine 302 resets the qubits in such channels being tested (e.g., channels 401A, 401B) by setting the qubits to the ground state in preparation for setting their excited states in the next test, as shown in units 602A-602B in Figure 6.

[0146] In step 904, with respect to the recently calibrated channel 401 (e.g., channel 401A), the test engine 302 determines the kernel response calculation using the read response (see step 902) and the trained kernel (kernel state of the recently calibrated channel 401).

[0147] As described above, in one embodiment, the test engine 302 uses the trained kernel to determine the kernel response computation (obtained in step 902).

[0148] In step 905, the test engine 302 compares the kernel response calculation (obtained in step 904) with the kernel response calculation obtained from the calibrated channel 401 (e.g., channel 401A) described above.

[0149] In step 906, the test engine 302 determines whether the difference between such kernel response calculations exceeds a threshold that may be user-specified.

[0150] As described above, such comparisons are performed to determine whether there is a threshold effect; that is, to determine whether the difference between such kernel response calculations exceeds a threshold; in other words, to determine whether the kernel response meets or fails to meet a user-defined criterion. As used herein, "criterion" means that the difference between kernel response calculations is less than a user-defined threshold.

[0151] If the difference exceeds a threshold (i.e., there is a threshold effect), in step 907, the test engine 301 shifts the phase of the signal on the most recently calibrated channel 401 (e.g., channel 401A). In one embodiment, the amount of the phase shift is a user-specified amount for each iteration. Such a phase shift of the signal may include adjusting the previously shifted phase, if the signal phase had been shifted previously.

[0152] In step 908, the calibration engine 301 recalibrates channel 401 (e.g., channel 401A) which has been calibrated using the phase-shifted signal as described above.

[0153] However, if the difference does not exceed the threshold (i.e., there is no threshold effect), in step 909, the test engine 302 does not shift or adjust the phase of the signal of channel 401 that has completed calibration (i.e., it does not adjust the phase of the signal that was previously shifted in the previous iteration).

[0154] In step 910, the state of the qubit is measured using a signal that includes a previously phase-shifted signal.

[0155] The process described above continues until the signal phase shift achieves the desired result (i.e., the difference between kernel response calculations is below a threshold). If such a desired result cannot be achieved, an error is reported, for example, to the user of classical computer 102.

[0156] In this way, the phase of the signal used to measure the qubit state is shifted, thereby eliminating the effects of noise when identifying the qubit state and increasing the speed and accuracy of state identification, thus improving the identification of qubit states in superconducting quantum computers.

[0157] Furthermore, the principles of this disclosure improve technologies or technologies related to superconducting quantum computing.

[0158] As described above, in superconducting quantum computers, qubits such as transmon qubits are controlled via the use of microwave radio frequency signals. That is, such transmon qubits can be manipulated by microwave radio frequency signals. The state of such transmon qubits is then read, for example, via the use of a microwave resonator, where the resonator's resonant frequency is shifted dispersively by the qubit state. As the number of qubits increases, the useful frequency range becomes crowded and noisy, and effective computation diminishes. In particular, it becomes difficult to identify the transmon qubit state using dynamic circuits, especially in kernel threshold systems. Dynamic circuits are quantum circuits that include circuit intermediate measurements, also known as feedforward operations, which affect the control flow of subsequent gate executions in the circuit. A standard and simplest example is a conditional reset, where a projection measurement of a qubit is performed. If the result is |1>, the state is flipped from the |1> state to the |0> state using an X gate. If the result is |0>, the state is not flipped. Dynamic circuits utilize a low-latency decision architecture known as a "kernel," which performs evaluations based on the state read from a qubit. The "kernel" refers to a time-domain filter calibrated for a specific task. Real-time convolution of the arriving signal (e.g., a microwave radio frequency signal) is then performed on the kernel to generate a combined result (also called the "kernel response") that is compared against a threshold to determine the quantum state of the qubit (e.g., a transmon qubit). That is, the read signal and the kernel generate a combined result ("kernel response") that is compared against a threshold to determine the quantum state of the qubit (e.g., a transmon qubit). For example, if the combined result exceeds the threshold, the result may be considered 1; if the combined result is below the threshold, the result may be considered 0. When computations involving such dynamic circuits are extended to other states, such as |2> and |3> states, in addition to |0> and |1> states, it becomes difficult to maintain the low computational latency required to enable proper operation.As a result, a fast decision-making kernel is applied to the sampled data in real time. Such an output is compared to a set of thresholds to determine the state readout. Unfortunately, kernel thresholding systems require a sufficiently large sample size for the kernel to clearly distinguish the input states. For example, the sample size can range from thousands to tens of thousands or even more. In latency-sensitive applications, such a large sample size can result in a long capture window, thereby slowing down the responsiveness of the dynamic circuit. Furthermore, kernel thresholding systems are often susceptible to in-band noise (e.g., crosstalk, Nyquist aliasing). Crosstalk is a disturbance caused by the electric or magnetic field of a signal affecting an adjacent signal. Nyquist aliasing refers to noise and signal images that are folded back from the upper Nyquist band to the first Nyquist band. As a result of such in-band noise, more samples are required, thereby further increasing the capture window and further slowing down the responsiveness of the dynamic circuit. Moreover, such noise can consequently make it difficult to identify qubit states (e.g., transmon qubit states). In addition, in multiplexed applications, bleed-over energy (energy spread across adjacent channels) can introduce dynamic decision-making problems in a similar manner to crosstalk, including difficulties in identifying qubit states (e.g., transmonqubit states). As a result, there is considerable interest in solutions to address the difficulty of identifying qubit states (e.g., transmonqubit states) in superconducting quantum computers, particularly those using dynamic circuits in kernel threshold systems.

[0159] Embodiments of the present disclosure improve such techniques by calibrating a channel (qubit readout channel) and training a kernel to include precise calibration data (also referred to herein as “kernel states”) used with the readout signal to obtain kernel response calculations. After calibrating the channel, a test is performed in which quantum operations are simultaneously performed on qubits in adjacent channels, including the recently calibrated channel, to determine whether the kernel response differs from the expected kernel response, where the expected kernel response exceeds a threshold based on the kernel states of the trained kernel of the recently calibrated channel. If such a situation occurs, the phase of the signal for the recently calibrated channel is shifted, and the process of recalibrating and testing the channel (using the phase-shifted signal) is repeated until the difference between the kernel response and the expected kernel response no longer exceeds a threshold. If such a situation occurs, the range of calculations for the kernel response is closest to the resonance point (the point where the signal frequency matches the natural frequency of the system), which effectively removes the effects of noise when identifying qubit states and improves the speed and accuracy of state identification. Furthermore, in this way there are improvements in the field of technology involving superconducting quantum computing.

[0160] The technical solutions provided by this disclosure cannot be implemented in the human mind or by a person using pen and paper. That is, the technical solutions provided by this disclosure cannot be implemented in the human mind or by a person using pen and paper without the use of a computer, in any reasonable amount of time, and with any reasonable degree of prediction.

[0161] In one embodiment of the present disclosure, a method for improving the identification of qubit states in a superconducting quantum computer comprises the step of shifting the phase of a signal in response to a kernel response not meeting a criterion. The method further comprises the step of using the phase-shifted signal to measure the state of a first qubit in response to a kernel response meeting a criterion.

[0162] In addition, in one embodiment of the present disclosure, the method further comprises a step of continuing to adjust the phase of the phase-shifted signal until the kernel response satisfies a criterion.

[0163] Furthermore, in one embodiment of the present disclosure, the method additionally includes a step of calibrating each of the multiple channels used to communicate the measured state of the qubit.

[0164] In addition, in one embodiment of the present disclosure, the method further comprises reading a plurality of responses of the second qubit set to the first quantum state for a first channel of a plurality of channels, as a result of a dynamic circuit performing a quantum operation on the second qubit set to the first quantum state.

[0165] Furthermore, in one embodiment of the present disclosure, the method further comprises the step of generating and recording a kernel state to be stored in the kernel for the first and second quantum states, based on the read response of a second qubit set to the first and second quantum states, in response to the completion of calibration of the first channel, wherein the kernel response is obtained from the calibrated first channel using the kernel state.

[0166] In addition, in one embodiment of the present disclosure, the method comprises the step of setting the excited state of a third qubit to a certain quantum state. The method further comprises the step of reading the response of the third qubit as a result of a dynamic circuit performing a quantum operation on the third qubit set to a certain quantum state. Furthermore, the method comprises the step of determining the kernel response of the third qubit based on the read response and kernel state of the third qubit. In addition, the method comprises the step of comparing the kernel response of the third qubit with a kernel response obtained from a calibrated first channel. Furthermore, the method comprises the step of shifting the phase of the signal of the calibrated first channel in response to the difference between the kernel response of the third qubit and the kernel response obtained from the calibrated first channel exceeding a threshold.

[0167] Furthermore, in one embodiment of the present disclosure, the method comprises the step of recalibrating the first channel using a phase-shifted signal.

[0168] Other embodiments of the methods described above include systems and computer program products.

[0169] The descriptions of the various embodiments of this disclosure are presented for illustrative purposes only and are not intended to be comprehensive or limitless to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein has been selected to best describe the principles of the embodiments, their practical applications, or technical improvements to the technology available on the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for improving the identification of qubit states in a superconducting quantum computer, The step of shifting the phase of the signal in response to the kernel response not meeting the criteria; and Steps to measure the state of the first qubit using the phase-shifted signal in response that the kernel response satisfies the criterion using the phase-shifted signal. A method that includes [a certain feature].

2. The step of continuing to adjust the phase of the phase-shifted signal until the kernel response satisfies the criterion. The method according to claim 1, further comprising:

3. The stage of calibrating each of the multiple channels used to communicate the measured state of the qubit. The method according to any of the prior claims, further comprising the above.

4. A step of reading a plurality of responses of the second qubit set to the first quantum state for a first channel of the plurality of channels, as a result of the dynamic circuit performing a quantum operation on the second qubit set to the first quantum state; and The step of reading multiple responses of the second qubit set to the second quantum state for the first channel among the multiple channels, as a result of the dynamic circuit performing a quantum operation on the second qubit set to the second quantum state. The method according to claim 3, further comprising:

5. In response to the completion of calibration of the first channel, a kernel state is generated and recorded for the first and second quantum states and stored in the kernel, based on the read response of the second qubit set to the first and second quantum states, wherein the kernel response is obtained from the calibrated first channel using the kernel state. The method according to claim 4, further comprising:

6. The step of setting the excited state of the third qubit to a certain quantum state; The step of reading the response of the third qubit as a result of the dynamic circuit performing a quantum operation on the third qubit which has been set to a certain quantum state; A step of determining the kernel response of the third qubit based on the read response of the third qubit and the kernel state; A step of comparing the kernel response of the third qubit with the kernel response obtained from the calibrated first channel; and Steps to shift the phase of the signal of the calibrated first channel in response to the difference between the kernel response of the third qubit and the kernel response obtained from the calibrated first channel exceeding a threshold. The method according to claim 5, further comprising:

7. The step of recalibrating the first channel using the phase-shifted signal. The method according to claim 6, further comprising:

8. A computer program product for improving the identification of qubit states in a superconducting quantum computer, wherein the computer program product comprises one or more computer-readable storage media on which the program code is embodied, and the program code is A procedure for shifting the phase of a signal in response to the kernel response not meeting a criterion; and A procedure for measuring the state of the first qubit using the phase-shifted signal in response that the kernel response satisfies the criterion using the phase-shifted signal. A computer program product that includes programming instructions for use.

9. The aforementioned program code is: A procedure to continue adjusting the phase of the phase-shifted signal until the kernel response satisfies the criterion. The computer program product according to claim 8, further comprising programming instructions for the purpose of the computer program.

10. The aforementioned program code is: Procedure for calibrating each of the multiple channels used to communicate the measured state of the qubit. A computer program product according to any one of claims 8 to 9, further comprising the programming instructions for the purposes of the computer program.

11. The aforementioned program code is: A procedure for reading out a plurality of responses of the second qubit set to the first quantum state for a first channel of the plurality of channels, as a result of a dynamic circuit performing a quantum operation on the second qubit set to the first quantum state; and A procedure for reading out multiple responses of the second qubit set to the second quantum state for the first channel among the multiple channels, as a result of the dynamic circuit performing a quantum operation on the second qubit set to the second quantum state. The computer program product according to claim 10, further comprising the programming instructions for the purposes of the computer program product according to claim 10.

12. The aforementioned program code is: In response to the completion of calibration of the first channel, a procedure for generating and recording kernel states for the first and second quantum states and storing them in a kernel, based on the read responses of the second qubit set to the first and second quantum states, wherein the kernel response is obtained from the calibrated first channel using the kernel state. The computer program product according to claim 11, further comprising the programming instructions for the purposes of the computer program product according to claim 11.

13. The aforementioned program code is: A procedure for setting the excited state of a third qubit to a certain quantum state; A procedure for reading out the response of the third qubit as a result of a dynamic circuit performing a quantum operation on the third qubit set to a certain quantum state; A procedure for determining the kernel response of the third qubit based on the read response of the third qubit and the kernel state; A procedure for comparing the kernel response of the third qubit with the kernel response obtained from the calibrated first channel; and A procedure for shifting the phase of the signal of the calibrated first channel in response to the difference between the kernel response of the third qubit and the kernel response obtained from the calibrated first channel exceeding a threshold. The computer program product according to claim 12, further comprising the programming instructions for the purposes of the computer program product according to claim 12.

14. The aforementioned program code is: The first channel is recalibrated using the phase-shifted signal. The computer program product according to claim 13, further comprising the programming instructions for the purpose of...

15. Memory for storing computer programs to improve the identification of qubit states in superconducting quantum computers; and A processor connected to the memory, wherein the processor is A procedure for shifting the phase of a signal in response to the kernel response not meeting a criterion; and A procedure for measuring the state of the first qubit using the phase-shifted signal in response that the kernel response satisfies the criterion using the phase-shifted signal. The computer program is configured to execute program instructions, including those of the computer program. A system equipped with these features.

16. The program instructions of the computer program are A procedure to continue adjusting the phase of the phase-shifted signal until the kernel response satisfies the criterion. The system according to claim 15, further comprising:

17. The program instructions of the computer program are Procedure for calibrating each of the multiple channels used to communicate the measured state of the qubit. The system according to any one of claims 15 to 16, further comprising:

18. The program instructions of the computer program are A procedure for reading out a plurality of responses of the second qubit set to the first quantum state for a first channel of the plurality of channels, as a result of a dynamic circuit performing a quantum operation on the second qubit set to the first quantum state; and A procedure for reading out multiple responses of the second qubit set to the second quantum state for the first channel among the multiple channels, as a result of the dynamic circuit performing a quantum operation on the second qubit set to the second quantum state. The system according to claim 17, further comprising:

19. The program instructions of the computer program are In response to the completion of calibration of the first channel, a procedure for generating and recording kernel states for the first and second quantum states and storing them in a kernel, based on the read responses of the second qubit set to the first and second quantum states, wherein the kernel response is obtained from the calibrated first channel using the kernel state. The system according to claim 18, further comprising:

20. The program instructions of the computer program are A procedure for setting the excited state of a third qubit to a certain quantum state; A procedure for reading the response of a third qubit as a result of a dynamic circuit performing a quantum operation on the third qubit which has been set to a certain quantum state. A procedure for determining the kernel response of the third qubit based on the read response of the third qubit and the kernel state; A procedure for comparing the kernel response of the third qubit with the kernel response obtained from the calibrated first channel; and A procedure for shifting the phase of the signal of the calibrated first channel in response to the difference between the kernel response of the third qubit and the kernel response obtained from the calibrated first channel exceeding a threshold. The system according to claim 19, further comprising:

Images