CN113052317B - Quantum state information acquisition method and device, quantum measurement and control system and computer - Google Patents
Quantum state information acquisition method and device, quantum measurement and control system and computer Download PDFInfo
- Publication number
- CN113052317B CN113052317B CN202110254067.8A CN202110254067A CN113052317B CN 113052317 B CN113052317 B CN 113052317B CN 202110254067 A CN202110254067 A CN 202110254067A CN 113052317 B CN113052317 B CN 113052317B
- Authority
- CN
- China
- Prior art keywords
- signal
- quantum
- model
- parameters
- quantum state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N10/00—Quantum computing, i.e. information processing based on quantum-mechanical phenomena
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a method and a device for acquiring quantum state information, a quantum measurement and control system and a computer. The IQ demodulation operation is not required to be carried out on the analog signals, and compared with the prior art, the scheme of the whole resolution method is simpler, and the calculation resources of the calculation module in the measurement and control system are effectively saved.
Description
Technical Field
The present invention relates to the field of quantum measurement and control, and in particular, to a method and apparatus for obtaining quantum state information, a quantum measurement and control system, and a computer.
Background
The qubit information refers to a quantum state of a qubit, the basic quantum states are a |0> state and a |1> state, after the qubit is operated, the quantum state of the qubit is changed, and on the quantum chip, the execution result of the quantum state of the qubit, namely the execution result of the quantum chip, is reflected after the quantum chip is executed, and the execution result is carried and transmitted by an acquired qubit reading signal (generally an analog signal). The rapid analysis of the quantum state of a qubit by a qubit read signal is a key task to understand the performance of a quantum chip. In the prior art, the resolution process of the qubit read signal can be generally divided into IQ demodulation, filtering and integration, and the resolution scheme needs to calculate the IQ demodulation result at the same time.
The inventor finds that in the prior art, the analysis scheme needs to calculate the IQ demodulation result at the same time, and after the calculation is completed, the IQ demodulation result needs to be rotated and the threshold value is judged, and the processes are complicated and need to consume a large amount of calculation resources of a calculation module in the measurement and control system.
Therefore, how to save computing resources is a technical problem to be solved in the art.
It should be noted that the information disclosed in the background section of the present application is only for enhancement of understanding of the general background of the present application and should not be taken as an admission or any form of suggestion that this information forms the prior art already known to those skilled in the art.
Disclosure of Invention
The application aims to provide a method and a device for acquiring quantum state information, a quantum measurement and control system and a computer, which are used for solving the problem that the existing analysis scheme needs to consume a large amount of calculation resources of a calculation module in the measurement and control system.
In order to solve the above technical problem, the present application provides a method for acquiring quantum state information, wherein the quantum state information is contained in an acquisition signal acquired from a qubit, the method comprising the steps of:
acquiring a first signal and a resolution coefficient corresponding to the first signal, wherein the first signal is a signal obtained by discretizing the acquired signal, and the resolution coefficient is determined according to related parameters when quantum states are extracted and determined from the acquired signal;
And analyzing the first signal by using a model to acquire quantum state information in the acquired signal, wherein the model is obtained by using a plurality of groups of data through machine learning training, each group of data in the plurality of groups of data comprises a second signal, a resolution coefficient and a label for identifying the quantum state of the second signal, and the second signal is obtained by discretizing the acquired signal for training.
Optionally, the relevant parameters include demodulation parameters and state classification equations, where the demodulation parameters are parameters configured in a demodulation process of the acquired signals, and the state classification equations are preconfigured to distinguish different quantum states.
Optionally, the process of obtaining the model through machine learning training using multiple sets of data includes:
acquiring a corresponding resolution coefficient and parameters of the state classification equation based on the second signal;
initializing the weight of the model based on the resolution coefficient and the parameters of the state classification equation;
based on the second signal and the weight, obtaining an output result of the model;
judging whether the output result meets a preset threshold condition or not;
If yes, stopping training, and outputting the trained model;
if not, acquiring a loss function based on the output result;
and updating the weight based on the loss function, and returning to execute the output result based on the second signal and the weight to acquire the model.
Optionally, the model comprises a nonlinear neuron.
Optionally, the nonlinear neuron comprises a Sigmoid neuron, or a Tanh neuron, or a Relu neuron.
Optionally, the loss function is obtained by the following formula:
wherein D is the second signal, td is the predicted result, od is the output result, and E is the loss function.
Alternatively, the weights are updated by a gradient descent method, or newton method, or random walk method, or evolutionary strategy method.
Optionally, the state classification equation is obtained by:
preparing a quantum bit into a first quantum state, repeatedly measuring the first quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a first set;
preparing the quantum bit into a second quantum state, and repeatedly measuring the second quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a second set, wherein: the first quantum state and the second quantum state are known quantum states and are different from each other;
The state classification equation is obtained based on the first set, the second set, and the orthogonal plane coordinate system.
Optionally, the state classification equation is obtained through a binary classification algorithm based on the first set, the second set and the orthogonal plane coordinate system.
Optionally, the demodulation parameter includes a quadrature local oscillator signal, where the quadrature local oscillator signal is configured to down-convert the acquisition signal and output a baseband signal.
Optionally, the demodulation parameters further include tap coefficients of a filter, where the filter is configured to perform filtering processing on the baseband signal.
Optionally, the demodulation parameters further include a window function, where the window function is a parameter configured in a process of performing weighted accumulation on the baseband signal after the filtering process.
Optionally, the model includes a quantum state resolution coefficient model, the quantum state resolution coefficient model includes a resolution equation formed by the resolution coefficients, and the resolution coefficients and the resolution equation are obtained according to the following formula:
wherein a, b, c are parameters of the state classification equation, cos (2πft l )、sin(2πft l ) For the orthogonal local oscillator signal, w 1(l+n) 、w 2(l+n) B for the window function n For the tap coefficient of the filter, N is the order of the filter, f is the frequency of the acquisition signal, s l In order to obtain a signal obtained by discretizing the acquired signal, L is the number of sampling points for discretizing the acquired signal, and k l For the resolution factor, t l And p is the calculation result of the resolution equation for the time point corresponding to each sampling point.
Based on the same inventive concept, the present invention also proposes an acquisition device of quantum state information, the quantum state information being contained in an acquisition signal acquired from a qubit, the acquisition device comprising:
the first module is configured to acquire a first signal and a corresponding resolution coefficient thereof, wherein the first signal is a signal obtained by discretizing the acquired signal, and the resolution coefficient is determined according to related parameters when quantum states are extracted and determined from the acquired signal;
the second module is configured to analyze the first signal by using a model to acquire quantum state information in the acquired signal, wherein the model is obtained by using a plurality of groups of data through machine learning training, each group of data in the plurality of groups of data comprises a second signal, a resolution coefficient and a label for identifying the quantum state of the second signal, and the second signal is a signal obtained by discretizing the acquired signal for training.
Based on the same inventive concept, the invention also provides a quantum state information acquisition system, which comprises:
the sampling unit is configured to sample the acquired signal containing the quantum state information and output a sampled signal;
the first unit is configured to store a model to analyze the sampling signal and obtain quantum state information in the acquisition signal, wherein the model is obtained by machine learning training through multiple groups of data, each group of data in the multiple groups of data comprises a second signal, a resolution coefficient and a label for identifying the quantum state of the second signal, and the second signal is obtained by discretizing the acquisition signal for training.
Based on the same inventive concept, the invention also provides a quantum measurement and control system, which comprises the quantum state information acquisition device or the quantum state information acquisition system;
the acquisition device or the acquisition system is arranged on a readout signal output side of the quantum measurement and control system, the readout signal output side is used for acquiring an acquisition signal output by the quantum measurement and control system, and the acquisition device or the acquisition system is used for distinguishing a quantum state corresponding to the acquisition signal.
Based on the same inventive concept, the invention also provides a quantum computer, which comprises the quantum measurement and control system.
Based on the same inventive concept, the invention further provides a readable storage medium, on which a computer program is stored, which when being executed by a processor, can implement the method for acquiring quantum state information in any of the above feature descriptions.
Compared with the prior art, the invention has the following beneficial effects:
according to the quantum state information acquisition method, first, the first signal and the corresponding resolution coefficient are acquired, a model is obtained through machine learning training, and the first signal is analyzed by the model. The IQ demodulation operation is not required to be carried out on the analog signals, and compared with the prior art, the scheme of the whole resolution method is simpler, and the calculation resources of the calculation module in the measurement and control system are effectively saved.
The quantum state information acquisition device, the quantum state information acquisition system, the quantum measurement and control system, the quantum computer and the readable storage medium provided by the invention belong to the same conception as the quantum state information acquisition method, so that the quantum state information acquisition device and the quantum state information acquisition system have the same beneficial effects and are not described in detail herein.
Drawings
Fig. 1 is a flow chart of a method for obtaining quantum state information according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a state classification equation according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a model using linear neurons according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a model using nonlinear neurons according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a linear neuron application in a scenario involving only linear factors;
FIG. 6 is a schematic diagram of a linear neuron applied in a scene containing nonlinear factors;
FIG. 7 is a graph showing the change of the loss function with the number of training times when training is performed using the model of FIG. 4.
Detailed Description
Specific embodiments of the present invention will be described in more detail below with reference to the drawings. Advantages and features of the invention will become more apparent from the following description and claims. It should be noted that the drawings are in a very simplified form and are all to a non-precise scale, merely for convenience and clarity in aiding in the description of embodiments of the invention.
In the description of the present invention, it should be understood that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", etc., are based on the directions or positional relationships shown in the drawings, are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
The artificial neural network can be traced back to the neuron model proposed by McCulloch-Pitts (M-P) in 1943 at the earliest. Rosenblatt adds a training process on the basis of M-P neurons, thereby proposing a perceptron model. So far, the artificial neural network has not only a perfect theoretical basis, but also plays an important role in practical application, and the artificial neural network covers the fields of pattern recognition, classification problems, multivariate data analysis and the like. As can be seen from the description of the schemes related to how to resolve the quantum states in the prior art, all the resolving schemes in the prior art need to be executed in a computing module in a measurement and control system, and various computing operations are performed step by the computing module, for example, the demodulation process is divided into steps of mixing, filtering, integrating, and the like, after demodulation is completed, a corresponding set of quadrature baseband components are generally output, and then the quantum states represented by the quadrature baseband components are judged. The steps are complicated, the calculation modules are required to be sequentially processed through a multi-stage pipeline, and the general calculation modules are processing chips with data processing functions, so that the calculation capacity of the processing chips is required to be fully considered during device type selection, and certain requirements are met on the development progress and cost of research and development personnel. The inventor finds that the specific step of carrying out quantum state resolution on the acquired signals containing the quantum state information can be carried out by a sensor based on an artificial neural network, namely, a model for acquiring the quantum state information is established, and when the method is applied specifically, the acquisition process of the quantum state information can be realized by inputting some data into the model for training and then inputting the acquired signals into the trained model. Therefore, the calculation module only needs to execute the calculation of the model and the acquired signals, and the quantum state information in the acquired signals can be acquired only by executing the operation in one step.
Referring to fig. 1, the present embodiment provides a method for acquiring quantum state information, where the quantum state information is contained in an acquisition signal acquired from a qubit, the method includes the following steps:
s1: acquiring a first signal and a resolution coefficient corresponding to the first signal, wherein the first signal is a signal obtained by discretizing the acquired signal, and the resolution coefficient is determined according to related parameters when quantum states are extracted and determined from the acquired signal;
s2: and analyzing the first signal by using a model to acquire quantum state information in the acquired signal, wherein the model is obtained by using a plurality of groups of data through machine learning training, each group of data in the plurality of groups of data comprises a second signal, a resolution coefficient and a label for identifying the quantum state of the second signal, and the second signal is obtained by discretizing the acquired signal for training.
The method for acquiring quantum state information provided by the embodiment is different from the prior art in that first signals and corresponding resolution coefficients are acquired, a model is obtained through machine learning training, and the model is used for analyzing the first signals. The IQ demodulation operation is not required to be carried out on the analog signals, and compared with the prior art, the scheme of the whole resolution method is simpler, and the calculation resources of the calculation module in the measurement and control system are effectively saved. The model can refer to fig. 3, the input of the model is the resolution coefficient and the first signal, the output of the model is the quantum state signal carried in the first signal, x in fig. 3 1 、x 2 ...x n Refer to the first signal, ω 0 、ω 1 、ω 2 ...ω n Refer to the resolution factor.
Specifically, the relevant parameters include demodulation parameters and state classification equations, wherein the demodulation parameters are parameters configured in the demodulation process of the acquired signals, and the state classification equations are parameters configured in advance for distinguishing different quantum states. In the embodiment of the present application, for convenience of description and understanding of the technical solution of the present application, different quantum states are the |0> state and the |1> state, respectively, it can be understood that other quantum states may be provided in other embodiments, and the present application is not limited herein.
Optionally, the state classification equation is obtained by:
the first step: preparing a quantum bit into a first quantum state, repeatedly measuring the first quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a first set;
and a second step of: preparing the quantum bit into a second quantum state, and repeatedly measuring the second quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a second set, wherein: the first quantum state and the second quantum state are known quantum states and are different from each other;
And a third step of: the state classification equation is obtained based on the first set, the second set, and the orthogonal plane coordinate system.
In the process of the quantum state information acquisition method, the signal to be processed is a microwave signal transmitted or reflected from the reading cavity, and the microwave signal is distinguished to carry quantum state information. The information of the quantum state is contained in the amplitude and phase of a certain frequency component, the frequency value is determined by the structure of the qubit of the experimental design, and for example, the frequency value can be set to be 500MHz-700MHz in the present embodiment. The parameter acquisition flow of the state classification equation is as follows: with reference to fig. 2, two sets of M (I, Q) will respectively form two-dimensional gaussian distributions (i.e. the first set and the second set) with different centers on the IQ plane according to the difference of the bit states, and we can obtain a corresponding optimal classification straight line by means of a general binary classification algorithm, as shown by the black dashed line in fig. 2, where we can assume that the state classification equation obtained by fig. 2 is ai+bq+c=0.
The sign of the output result of the model can be directly used for distinguishing quantum states, a single-multiplication accumulator fast distinguishing channel based on the FPGA in particular in a measurement and control system can directly take the highest bit, namely the sign bit, of the binary signed value of the output result of the model, and 0/1 can directly correspond to two quantum states. For example, assuming that the graph of the state classification equation is shown in fig. 2, since the highest order of the calculation result of the resolution equation represents positive and negative thereof, the highest order represents a negative number when 1, and the highest order represents a positive number when 0. Therefore, when the highest order is 1, it indicates that the calculation result is below the dashed line in fig. 2, the corresponding quantum state is the |0> state, and when the highest order is 0, it indicates that the calculation result is above the dashed line in fig. 2, the corresponding quantum state should be the |1> state.
It can be understood that in other embodiments, the output result of the model may be compared with a preset threshold, and in fact, in this embodiment, the threshold may be understood as 0, and the technical scheme for setting the threshold is similar to that of the present application and will not be described herein.
Optionally, the state classification equation is obtained through a binary classification algorithm based on the first set, the second set and the orthogonal plane coordinate system.
Optionally, the demodulation parameter includes a quadrature local oscillator signal, where the quadrature local oscillator signal is configured to down-convert the acquisition signal and output a baseband signal. Typically, the quadrature local oscillator signal is cos (2pi ft l ) Sum sin (2 pi ft) l )。
Further, since the baseband signal output after down-converting the acquisition signal contains a high-frequency noise signal that does not contain sub-bit information, the baseband signal output after down-converting needs to be filtered, and therefore, when designing the demodulation parameter, the tap coefficient of the filter needs to be taken into consideration, that is, the demodulation parameter further includes the tap coefficient of the filter, and the filter is used for filtering the baseband signal. It will be appreciated that the filter may be a low-pass filter, where the tap coefficients of the filter are related to the order of the low-pass filter, and in general, the low-pass filter of the order N has n+1 tap coefficients, and the values of the specific tap coefficients may be designed according to the actual situation, which is not limited herein.
Furthermore, the baseband signals after filtering can be directly and respectively subjected to parallel summation and de-averaging to be used as demodulation results, and then quantum state resolution is carried out by using the demodulation results. The inventors have found that the fidelity of the quantum state resolution can be optimized to some extent by designing the window function to implement weighted accumulation instead of a general rectangle. Therefore, when designing the demodulation coefficients, parameters of a window function may also be taken into account, that is, the demodulation parameters further include a window function, where the window function is a parameter configured in the process of performing weighted accumulation on the baseband signal after the filtering process.
For better understanding of the technical solution of the present application, the following briefly describes the related concepts of the window function: the main mathematical tool for digital signal processing is the fourier transform, which studies the relation between the whole time domain and the frequency domain. However, when engineering test signal processing is implemented using a computer, it is not possible to measure and calculate an infinitely long signal, but rather take a finite time slice for analysis. The method comprises the steps of intercepting a time slice from a signal, performing cycle prolongation processing by using the intercepted signal time slice to obtain a virtual infinitely long signal, and performing mathematical processing such as Fourier transformation, correlation analysis and the like on the signal. After the infinitely long signal is truncated, the spectrum is distorted and the energy originally concentrated at f (0) is dispersed into two wider bands (this phenomenon is called spectral energy leakage). In order to reduce the leakage of spectral energy, the signal may be truncated by using different clipping functions, which are called window functions, which are signals with limited time domain. In the field of quantum signal processing, window function design methods more suitable for this specific problem have been developed, and the difference between normalized two-state separation trajectories (i.e., multiple averages of the difference between corresponding two-state sequences after filtering (note that IQ two paths are needed)) is generally adopted.
Optionally, the quantum state resolution coefficient model includes a resolution coefficient, the resolution model includes a resolution equation, and the resolution coefficient and the resolution equation are obtained according to the following formula:
wherein a, b, c are parameters of the state classification equation, cos (2πft l )、sin(2πft l ) For the orthogonal local oscillator signal, w 1(l+n) 、w 2(l+n) B for the window function n For the tap coefficient of the filter, N is the order of the filter, f is the frequency of the acquisition signal, s l In order to obtain a signal obtained by discretizing the acquired signal, L is the number of sampling points for discretizing the acquired signal, and k l For the resolution factor, t l And p is the calculation result of the resolution equation for the time point corresponding to each sampling point. It should be noted that ω in FIG. 3 0 And c is a parameter in the state classification equation.
In the following description of the present application with reference to a specific example, in this embodiment, it may be assumed that four data points in the acquired signal are {5,2, -1, -2}, and the corresponding two-sign sine wave (quadrature local oscillator signal) sampling sequences are {1,0, -1,0} and {0,1, -1,0}. The filter uses a first order filter with coefficients set to 0.5, 0.5. The two-way window function is designed as {0.2,0.2,0.3,0.3} and {0.2,0.3,0.3,0.2} respectively, the state classification equation is set as x+y-1=0, and the graph of the state classification equation can be shown with reference to fig. 2. The frequency of the analog signal may be set to 500MHz-700MHz, where 500MHz is optional. Substituting the above parameters into equation 2 to obtain the resolution coefficient {0.2,0.3, -0.55,0}, then inputting the sampled four data points {5,2, -1, -2} and the resolution coefficient {0.2,0.3, -0.55,0} into the model, please refer to fig. 3, namely, taking the sampled four data points {5,2, -1, -2} and the resolution coefficient {0.2,0.3, -0.55,0} and the parameter c of the state classification equation as input into the model, firstly, calculating the result by the resolution equation, wherein the result of the resolution equation is 1.15, referring to fig. 3, when the result of the resolution equation is 1.15, the output result o (x) of the model is 1, and when the output result of the model is 1, it can be seen according to fig. 2, the corresponding quantum state is above the straight line, so that the quantum state corresponding to the acquired signal is |1> -state can be resolved.
Specifically, the training process for the model is specifically as follows, that is, the process of obtaining the model through machine learning training by using multiple sets of data includes:
acquiring a corresponding resolution coefficient and parameters of the state classification equation based on the second signal;
initializing the weight of the model based on the resolution coefficient and the parameters of the state classification equation; the weights are also ω in FIG. 3 0 、ω 1 、ω 2 ...ω n Wherein ω is 0 Initializing parameters c, omega of the state classification equation 1 、ω 2 ...ω n Initializing to the resolution factor.
Based on the second signal and the weight, obtaining an output result of the model;
judging whether the output result meets a preset threshold condition or not;
if yes, stopping training, and outputting the trained model;
if not, acquiring a loss function based on the output result;
and updating the weight based on the loss function, and returning to execute the output result based on the second signal and the weight to acquire the model.
Specifically, the loss function is obtained by the following formula:
wherein D is the second signal, td is the predicted result, od is the output result, and E is the loss function.
It is noted that the weights may be updated by a gradient descent method, or newton method, or random walk method, or evolutionary strategy method, preferably by a gradient descent method. The inventors found when carrying out the implementation using the above scheme of the present embodiment: although the method for acquiring quantum state information provided in the embodiment saves the computing resources of the computing module in the measurement and control system, it actually uses a linear neuron, and when the output data only includes the linear factor, the linear neural network can smoothly complete classification, please refer to fig. 5. However, if the situation becomes complicated, the data becomes linearly inseparable, and the straight line cannot be well classified, see fig. 6. In fig. 5 and fig. 6, the triangle and the square represent two different quantum states, and comparing fig. 5 and fig. 6, it is not difficult to see that when the output data is not completely linear, effective quantum state information cannot be obtained by implementing the model of the linear neuron.
From the above description of the training process of the model, in connection with fig. 6, it is known that if linear neurons are used in the model, only a small adjustment of the weights, i.e. a small linear movement in fig. 6, will actually result in a huge classification change, and if it is desired to classify these things well, we actually need a rugged curve, i.e. only changing some points on the line, instead of moving the trajectory of the whole line for each change. Based on this, the inventor further found that, because the quantum bit reading signal includes part of noise with quantum characteristics and nonlinear factors of the quantum bit reading signal, when a group of quantum bit reading signals are analyzed by using a model of linear neurons, the fidelity of quantum state information which can be effectively obtained from the sampling signal can only reach 95% at most. In addition, as the effects of noise and nonlinear factors increase, the upper fidelity limit of quantum state resolution using a model of linear neurons continues to decrease.
Further, in order to solve the problem that the model of the linear neuron is insufficient in fidelity in the process of obtaining the quantum state information, the inventors consider introducing the nonlinear neuron into the model, please refer to fig. 4. The model includes nonlinear neurons. Specifically, the nonlinear neuron may be a Sigmoid neuron, a Tanh neuron, or a Relu neuron. In this embodiment, sigmoid neurons are taken as an example, and other types of neurons are similar to the Sigmoid neurons, and are not described in detail herein. The input of Sigmoid neuron is the same as the input of the front linear neuron, but its output is not used with the linear neuron, and its output result is not limited to only two fixed values, but can take all values between 0 and 1, say 0.972 is a valid input value.
It should be noted that, since the output of the model using Sigmoid neurons is all values between 0 and 1, it is necessary to preset a threshold according to the actual situation and distinguish two different quantum states according to the magnitude relation between the output result of the actual model and the threshold. For example, the threshold may be set to 0.5, with quantum states corresponding to greater than 0.5 being the |1> state and quantum states corresponding to less than or equal to 0.5 being the |0> state. In order to facilitate understanding of the technical solution of the present application, in this embodiment, the threshold values are all set to 0.5, and the quantum states corresponding to more than 0.5 are all |1> states, and the quantum states corresponding to less than or equal to 0.5 are all |0> states as the judgment criteria. For example, when the result of the data input to the model is 0.73 after passing through the Sigmoid neuron, the result is greater than 0.5, and thus the corresponding quantum state is the |1> state. It will be appreciated that the thresholds for different types of nonlinear neurons are also different and need to be selected according to the actual situation, and are not limited herein.
In order to verify the difference of the fidelity of the quantum state information obtained effectively from the sampling signals by the linear neuron model and the nonlinear neuron model, the inventor firstly presets 10000 groups of independent measurement waveforms, wherein half of the signals are prepared in an |0> state in advance, and the other half of the signals are prepared in an |1> state, the frequency is set to 618.5MHz, and when the quantum state information is obtained according to the linear neuron model, the fidelity of the |0> state is 0.928, and the fidelity of the |1> state is 0.889. Then, quantum state information is obtained according to a model of a nonlinear neuron (here, a Sigmoid neuron is adopted), fig. 7 is a curve of a change of a loss function with training times when training is performed by using the model of the nonlinear neuron, the abscissa of fig. 7 is a training coefficient, the ordinate is a value of a loss function E, and the fidelity of the |0> state is 0.949 and the fidelity of the |1> state is 0.910 after training for the same times. It can be seen that the fidelity of the quantum state information obtained by using the model of the nonlinear neuron is obviously improved compared with that of the quantum state information obtained by using the model of the linear neuron.
By applying machine learning to the acquisition of quantum state information, machine learning training is performed on the basis of a model, and nonlinear neurons are introduced into the model, so that the model can effectively avoid the influence of noise of quantum characteristics and nonlinear factors of a quantum bit reading signal, the fidelity of the quantum state information effectively obtained from a sampling signal is effectively improved, in addition, the efficiency of acquiring the quantum bit reading signal can be effectively improved due to the adoption of the model trained by machine learning, and the realization efficiency of the whole acquisition method is improved to a certain extent.
It will be appreciated that the process of machine learning training with a model of nonlinear neurons is essentially the same as that described above, except that each minute change in weight causes a corresponding minute change in output and changes only some points on the line, and does not affect the trajectory of the entire line. Therefore, when the quantum state information is acquired by using the nonlinear neuron model, the fidelity of the quantum state information effectively acquired from the sampling signal can be effectively improved.
Based on the same inventive concept, the present embodiment further provides an acquisition apparatus of quantum state information, where the quantum state information is included in an acquisition signal acquired from a qubit, the acquisition apparatus including:
the first module is configured to acquire a first signal and a corresponding resolution coefficient thereof, wherein the first signal is a signal obtained by discretizing the acquired signal, and the resolution coefficient is determined according to related parameters when quantum states are extracted and determined from the acquired signal;
the second module is configured to analyze the first signal by using a model to acquire quantum state information in the acquired signal, wherein the model is obtained by using a plurality of groups of data through machine learning training, each group of data in the plurality of groups of data comprises a second signal, a resolution coefficient and a label for identifying the quantum state of the second signal, and the second signal is a signal obtained by discretizing the acquired signal for training.
It will be appreciated that the first module and the second module may be combined in one device or any one of them may be split into a plurality of sub-modules, or that at least part of the functions of one or more of the first module and the second module may be combined with at least part of the functions of the other modules and implemented in one functional module. According to embodiments of the invention, at least one of the first module and the second module may be implemented at least partially as hardware circuitry, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or any other reasonable way of integrating or packaging circuits, or as hardware or firmware, or as a suitable combination of software, hardware, and firmware implementations. Alternatively, at least one of the first module and the second module may be at least partially implemented as a computer program module, which when executed by a computer, may perform the functions of the respective module.
Based on the same inventive concept, the present embodiment further provides a system for acquiring quantum state information, including:
the sampling unit is configured to sample the acquired signal containing the quantum state information and output a sampled signal;
the first unit is configured to store a model to analyze the sampling signal and obtain quantum state information in the acquisition signal, wherein the model is obtained by machine learning training through multiple groups of data, each group of data in the multiple groups of data comprises a second signal, a resolution coefficient and a label for identifying the quantum state of the second signal, and the second signal is obtained by discretizing the acquisition signal for training.
Based on the same inventive concept, the embodiment also provides a quantum measurement and control system, which comprises the quantum state information acquisition device or the quantum state information acquisition system;
the acquisition device or the acquisition system is arranged on a readout signal output side of the quantum measurement and control system, the readout signal output side is used for acquiring an acquisition signal output by the quantum measurement and control system, and the acquisition device or the acquisition system is used for distinguishing a quantum state corresponding to the acquisition signal.
Based on the same inventive concept, the embodiment also provides a quantum computer, which comprises the quantum measurement and control system.
Based on the same inventive concept, the present embodiment also proposes a readable storage medium having stored thereon a computer program, which when executed by a processor, is capable of implementing the method for obtaining quantum state information according to any of the above feature descriptions.
The readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device, such as, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the preceding. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: portable computer disks, hard disks, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static Random Access Memory (SRAM), portable compact disk read-only memory (CD-ROM), digital Versatile Disks (DVD), memory sticks, floppy disks, mechanical coding devices, punch cards or in-groove structures such as punch cards or grooves having instructions stored thereon, and any suitable combination of the foregoing. The computer program described herein may be downloaded from a readable storage medium to a respective computing/processing device or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. The network interface card or network interface in each computing/processing device receives the computer program from the network and forwards the computer program for storage in a readable storage medium in the respective computing/processing device. Computer programs for carrying out operations of the present invention may be assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer program may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing electronic circuitry, such as programmable logic circuits, field Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), with state information for a computer program, which can execute computer-readable program instructions.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer programs. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the programs, when executed by the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer programs may also be stored in a readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the readable storage medium storing the computer program includes an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer program may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the computer program which is executed on the computer, other programmable apparatus or other devices implements the functions/acts specified in the flowchart and/or block diagram block or blocks.
In the description of the present specification, a description of the terms "one embodiment," "some embodiments," "examples," or "particular examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics may be combined in any suitable manner in any one or more embodiments. Further, one skilled in the art can engage and combine the different embodiments or examples described in this specification.
The foregoing is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any person skilled in the art will make any equivalent substitution or modification to the technical solution and technical content disclosed in the invention without departing from the scope of the technical solution of the invention, and the technical solution of the invention is not departing from the scope of the invention.
Claims (15)
1. A method of acquiring quantum state information contained in an acquisition signal acquired from a qubit, the method comprising the steps of:
Acquiring a first signal and a corresponding resolution coefficient thereof, wherein the first signal is a signal obtained after discretization processing is carried out on the acquired signal, the resolution coefficient is determined according to related parameters when quantum states are extracted and determined from the acquired signal, the related parameters comprise demodulation parameters and parameters of a state classification equation, the demodulation parameters are parameters configured in the demodulation process of the acquired signal, and the state classification equation is a parameter pre-configured for distinguishing different quantum states;
training a model for first signal analysis based on the acquired second signal and its corresponding resolution coefficient and parameters of the state classification equation, and a tag identifying a quantum state of the second signal, wherein the resolution coefficient and parameters of the state classification equation are used as initial values of weights of the model, and the weights are iteratively updated in a training process until a trained model is obtained; the second signal is a signal obtained after discretization processing is carried out on the acquired signal for training;
analyzing the first signal by using a trained model to acquire quantum state information in the acquired signal; the state classification equation is obtained through the following steps:
Preparing a quantum bit into a first quantum state, repeatedly measuring the first quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a first set;
preparing the quantum bit into a second quantum state, and repeatedly measuring the second quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a second set, wherein: the first quantum state and the second quantum state are known quantum states and are different from each other;
the state classification equation is obtained based on the first set, the second set, and the orthogonal plane coordinate system.
2. The method of claim 1, wherein the model for the first signal analysis is trained based on the acquired second signal and its corresponding resolution and parameters of the state classification equation, and a tag identifying the quantum state of the second signal, wherein the resolution and parameters of the state classification equation are used as initial values of weights of the model, and the weights are iteratively updated during training until a trained model is obtained; comprising the following steps:
acquiring a corresponding resolution coefficient and parameters of the state classification equation based on the second signal;
Initializing the weight of the model based on the resolution coefficient and the parameters of the state classification equation;
based on the second signal and the weight, obtaining an output result of the model;
judging whether the output result meets a preset threshold condition or not;
if yes, stopping training, and outputting the trained model;
if not, acquiring a loss function based on the output result;
and updating the weight based on the loss function, and returning to execute the output result based on the second signal and the weight to acquire the model.
3. The method of claim 2, wherein the model comprises nonlinear neurons.
4. A method of obtaining quantum state information as claimed in claim 3 wherein the nonlinear neuron comprises a Sigmoid neuron, or a Tanh neuron, or a Relu neuron.
5. The method of claim 2, wherein the loss function is obtained by the following formula:
wherein D is the second signal, t d O as a result of prediction d E is the loss function and d is a sub-term constituting the second signal for the output result.
6. The method of claim 2, wherein the weights are updated by a gradient descent method, a newton method, a random walk method, or an evolutionary strategy method.
7. The method of claim 1, wherein the state classification equation is obtained by a binary classification algorithm based on the first set, the second set, and the orthogonal plane coordinate system.
8. The method of claim 1, wherein the demodulation parameters include a quadrature local oscillator signal, the quadrature local oscillator signal being used to down-convert the acquisition signal and output a baseband signal.
9. The method of claim 8, wherein the demodulation parameters further comprise tap coefficients of a filter, the filter being configured to filter the baseband signal.
10. The method of claim 9, wherein the demodulation parameters further include a window function, the window function being parameters configured in a process of weighted accumulation of the baseband signals after the filtering process.
11. The method of claim 10, wherein the model includes a quantum state resolution coefficient model, the quantum state resolution coefficient model includes a resolution equation formed by the resolution coefficients, and the resolution coefficients and the resolution equation are obtained according to the following formula:
wherein a, b, c are parameters of the state classification equation, cos (2πft l )、sin(2πft l ) For the orthogonal local oscillator signal, w 1(l+n) 、w 2(l+n) B for the window function n For the tap coefficient of the filter, N is the order of the filter, f is the frequency of the acquisition signal, s l In order to obtain a signal obtained by discretizing the acquired signal, L is the number of sampling points for discretizing the acquired signal, and k l For the resolution factor, t l For each sampling point, p is the calculation result of the resolution equation, L is a natural number from 1 to L, and N is a natural number from 0 to N.
12. An acquisition apparatus of quantum state information, wherein the quantum state information is contained in an acquisition signal acquired from a qubit, the acquisition apparatus comprising:
the first module is used for acquiring a first signal and a resolution coefficient corresponding to the first signal, wherein the first signal is a signal obtained after discretization processing is carried out on the acquired signal, the resolution coefficient is determined according to related parameters when quantum states are extracted and determined from the acquired signal, the related parameters comprise demodulation parameters and parameters of a state classification equation, the demodulation parameters are parameters configured in the demodulation process of the acquired signal, and the state classification equation is a parameter pre-configured for distinguishing different quantum states;
The model training module is used for training a model for first signal analysis based on the acquired second signal, the corresponding resolution coefficient and the parameter of the state classification equation and the label for identifying the quantum state of the second signal, wherein the resolution coefficient and the parameter of the state classification equation are used as initial values of the weight of the model, and the weight is iteratively updated in the training process until a trained model is obtained; the second signal is a signal obtained after discretization processing is carried out on the acquired signal for training;
the second module is used for analyzing the first signal by using the trained model to acquire quantum state information in the acquired signal;
the state classification equation is obtained through the following steps:
preparing a quantum bit into a first quantum state, repeatedly measuring the first quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a first set;
preparing the quantum bit into a second quantum state, and repeatedly measuring the second quantum state to obtain a plurality of coordinate point data of a quantum bit reading signal on an orthogonal plane coordinate system, and marking the coordinate point data as a second set, wherein: the first quantum state and the second quantum state are known quantum states and are different from each other;
The state classification equation is obtained based on the first set, the second set, and the orthogonal plane coordinate system.
13. A quantum measurement and control system, comprising the quantum state information acquisition device of claim 12;
the acquisition device is arranged on a readout signal output side of the quantum measurement and control system, the readout signal output side is used for acquiring an acquisition signal output by the quantum measurement and control system, and the acquisition device or the acquisition system is used for distinguishing a quantum state corresponding to the acquisition signal.
14. A quantum computer comprising the quantum measurement and control system of claim 13.
15. A readable storage medium having stored thereon a computer program, which when executed by a processor is capable of implementing the method of obtaining quantum state information according to any of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110254067.8A CN113052317B (en) | 2021-03-09 | 2021-03-09 | Quantum state information acquisition method and device, quantum measurement and control system and computer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110254067.8A CN113052317B (en) | 2021-03-09 | 2021-03-09 | Quantum state information acquisition method and device, quantum measurement and control system and computer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113052317A CN113052317A (en) | 2021-06-29 |
| CN113052317B true CN113052317B (en) | 2023-10-13 |
Family
ID=76510538
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110254067.8A Active CN113052317B (en) | 2021-03-09 | 2021-03-09 | Quantum state information acquisition method and device, quantum measurement and control system and computer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113052317B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11606230B2 (en) | 2021-03-03 | 2023-03-14 | Apple Inc. | Channel equalization |
| US11784731B2 (en) * | 2021-03-09 | 2023-10-10 | Apple Inc. | Multi-phase-level signaling to improve data bandwidth over lossy channels |
| CN113807526B (en) * | 2021-09-26 | 2024-03-29 | 深圳市腾讯计算机***有限公司 | Method, device, equipment and storage medium for acquiring eigenstates of quantum system |
| CN116523058B (en) * | 2022-01-24 | 2024-07-16 | 本源量子计算科技（合肥）股份有限公司 | Quantum gate operation information acquisition method and device and quantum computer |
| CN115115055B (en) * | 2022-08-31 | 2022-12-06 | 合肥本源量子计算科技有限责任公司 | Parameter optimization method and device for combined read signal and quantum control system |
| CN115146781B (en) * | 2022-09-01 | 2022-12-06 | 合肥本源量子计算科技有限责任公司 | Parameter acquisition method and device for combined read signal and quantum control system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109447271A (en) * | 2018-10-15 | 2019-03-08 | 合肥本源量子计算科技有限责任公司 | A kind of quantum bit quantum state read method and device |
| CN110706254A (en) * | 2019-09-19 | 2020-01-17 | 浙江大学 | Target tracking template self-adaptive updating method |
| CN111311000A (en) * | 2020-02-14 | 2020-06-19 | 北京百度网讯科技有限公司 | User consumption behavior prediction model training method, device, equipment and storage medium |
| CN111369003A (en) * | 2020-03-31 | 2020-07-03 | 合肥本源量子计算科技有限责任公司 | Method and device for determining fidelity of quantum bit reading signal |
| CN111368920A (en) * | 2020-03-05 | 2020-07-03 | 中南大学 | Quantum twin neural network-based binary classification method and face recognition method thereof |
| CN111563186A (en) * | 2020-04-30 | 2020-08-21 | 北京百度网讯科技有限公司 | Quantum data storage method, quantum data reading method, quantum data storage device, quantum data reading device and computing equipment |
| CN112001498A (en) * | 2020-08-14 | 2020-11-27 | 苏州浪潮智能科技有限公司 | Data identification method and device based on quantum computer and readable storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6882533B2 (en) * | 2017-06-02 | 2021-06-02 | グーグル エルエルシーＧｏｏｇｌｅ ＬＬＣ | Quantum neural network |
| US10528885B2 (en) * | 2017-09-29 | 2020-01-07 | International Business Machines Corporation | Cross-resonance fan-out for efficiency and hardware reduction |
-
2021
- 2021-03-09 CN CN202110254067.8A patent/CN113052317B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109447271A (en) * | 2018-10-15 | 2019-03-08 | 合肥本源量子计算科技有限责任公司 | A kind of quantum bit quantum state read method and device |
| CN110706254A (en) * | 2019-09-19 | 2020-01-17 | 浙江大学 | Target tracking template self-adaptive updating method |
| CN111311000A (en) * | 2020-02-14 | 2020-06-19 | 北京百度网讯科技有限公司 | User consumption behavior prediction model training method, device, equipment and storage medium |
| CN111368920A (en) * | 2020-03-05 | 2020-07-03 | 中南大学 | Quantum twin neural network-based binary classification method and face recognition method thereof |
| CN111369003A (en) * | 2020-03-31 | 2020-07-03 | 合肥本源量子计算科技有限责任公司 | Method and device for determining fidelity of quantum bit reading signal |
| CN111563186A (en) * | 2020-04-30 | 2020-08-21 | 北京百度网讯科技有限公司 | Quantum data storage method, quantum data reading method, quantum data storage device, quantum data reading device and computing equipment |
| CN112001498A (en) * | 2020-08-14 | 2020-11-27 | 苏州浪潮智能科技有限公司 | Data identification method and device based on quantum computer and readable storage medium |
Non-Patent Citations (5)
| Title |
|---|
| Quantum pattern recognition;P. W. H. Pinkse et al.;《2013 Conference on Lasers & Electro-Optics Europe & International Quantum Electronics Conference CLEO EUROPE/IQEC》;第1页 * |
| Recognition Property of Quantum Character Recognition Algorithm;Osamu Matoba et al.;《 LEOS 2006 - 19th Annual Meeting of the IEEE Lasers and Electro-Optics Society》;第350-351页 * |
| 机器学习在量子信息与其他物理中的应用;崔子嵬;《中国优秀硕士学位论文全文数据库基础科学辑》(第第01期期);第A005-64页 * |
| 量子机器学习;陆思聪等;《控制理论与应用》;第34卷(第11期);第1429-1436页 * |
| 量子神经元特性研究;李飞等;《电路与***学报》(第04期);第76-80页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113052317A (en) | 2021-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113052317B (en) | Quantum state information acquisition method and device, quantum measurement and control system and computer | |
| Stanev et al. | Unsupervised phase mapping of X-ray diffraction data by nonnegative matrix factorization integrated with custom clustering | |
| Kolosnjaji et al. | Empowering convolutional networks for malware classification and analysis | |
| Ratto et al. | State dependent parameter metamodelling and sensitivity analysis | |
| Kumar et al. | Fractional Fourier transform and fractional-order calculus-based image edge detection | |
| Agarwal et al. | Towards the unification and robustness of perturbation and gradient based explanations | |
| Moini et al. | Power side-channel attacks on BNN accelerators in remote FPGAs | |
| Mamalakis et al. | Explainable artificial intelligence in meteorology and climate science: Model fine-tuning, calibrating trust and learning new science | |
| CN112884154B (en) | Quantum state resolution method, device and system, quantum measurement and control system and computer | |
| Greaves-Tunnell et al. | A statistical investigation of long memory in language and music | |
| Maiorino et al. | Data-driven detrending of nonstationary fractal time series with echo state networks | |
| US20240013077A1 (en) | Phase-robust matched kernel acquisition for qubit state determination | |
| Aftab et al. | Diagnosis of plant-wide oscillations by combining multivariate empirical mode decomposition and delay vector variance | |
| AU2018271286A1 (en) | Systems and methods for obtaining optimal mother wavelets for facilitating machine learning task | |
| Romor et al. | Multi‐fidelity data fusion for the approximation of scalar functions with low intrinsic dimensionality through active subspaces | |
| CN112116018A (en) | Sample classification method, apparatus, computer device, medium, and program product | |
| Taran et al. | Challenging interferometric imaging: Machine learning-based source localization from uv-plane observations | |
| Ferreira et al. | A model validation scale based on multiple indices | |
| Künkel et al. | Detecting pulsars with neural networks: a proof of concept | |
| Arrubarrena et al. | Novelty Detection on Radio Astronomy Data using Signatures | |
| Azad et al. | qLEET: visualizing loss landscapes, expressibility, entangling power and training trajectories for parameterized quantum circuits | |
| Murbach et al. | Open software and datasets for the analysis of electrochemical impedance spectra | |
| Sree et al. | Intelligent Malware detection using extreme learning machine | |
| Nose-Filho et al. | On disjoint component analysis | |
| Rahul et al. | Long Short-Term Memory-Convolution Neural Network Based Hybrid Deep Learning Approach for Power Quality Events Classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 230088 6th floor, E2 building, phase II, innovation industrial park, 2800 innovation Avenue, high tech Zone, Hefei City, Anhui Province Applicant after: Benyuan Quantum Computing Technology (Hefei) Co.,Ltd. Address before: 230088 6th floor, E2 building, phase II, innovation industrial park, 2800 innovation Avenue, high tech Zone, Hefei City, Anhui Province Applicant before: ORIGIN QUANTUM COMPUTING COMPANY, LIMITED, HEFEI |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |