15 results about "Amplitude amplification" patented technology

The invention belongs to the technical field of intelligent transportation, and particularly relates to an automatic driving automobile key scene generation method based on quantum reinforcement learning. Comprising the following steps: step 1, constructing a simulation environment; 2, designing a continuous control-oriented quantum strategy network; 3, constructing a multi-dimensional sparse reward function; 4, performing interactive training and optimization on the quantum strategy network; step 5, outputting a structured test case and performing visual analysis; according to the method, a key scene is generated and modeled as a dynamic interaction process of an intelligent agent and an environment, sparse dangerous behaviors are efficiently explored in a high-dimensional state space through a learnable quantum strategy network, the search efficiency of the sparse key scene in the high-dimensional scene space is remarkably improved by introducing a quantum superposition state and a quantum amplitude amplification mechanism, and the search efficiency of the sparse key scene in the high-dimensional scene space is improved. And the problems of combination explosion and sparsity faced by the traditional method are solved.

The application discloses a quantum state preparation method and device for non-uniform information, and the method comprises the following steps: firstly, obtaining non-uniform information; then, converting each element in a data vector corresponding to the non-uniform information into a first data and a second data; obtaining first quantum state information according to first information and a sub-quantum state preparation circuit; obtaining second quantum state information according to second information stored in a quantum random access circuit; and finally, evolving the first quantum state information and the second quantum state information based on an amplitude amplification quantum circuit to obtain a target quantum state containing the non-uniform information. The target quantum state is obtained by converting a data vector to be prepared, and constructing a sub-quantum state preparation circuit and an amplitude amplification quantum circuit, so that the number of times of amplitude amplification of quantum state preparation is reduced, and the depth and the number of operations of the quantum state preparation circuit for preparing non-uniform information are reduced.

The invention discloses a black box quantum state preparation method based on grouping processing. The method comprises the steps that according to a mark bit and an initial quantum state, a coarse proportion estimation value of a target state is determined, and the initial quantum state is a superposition state obtained by processing an input data vector based on a black box. Under the condition that the coarse proportion estimation value is smaller than a grouping threshold value, separation processing is conducted on the initial quantum state according to a first auxiliary bit, a first intermediate quantum state is determined, the grouping threshold value is determined based on the number of input data vectors, the first intermediate quantum state comprises a target state part and an orthogonal state part, the target state part comprises a first sub-quantum state, and the orthogonal state part comprises a second sub-quantum state; the orthogonal state portion includes a second sub-quantum state. And according to the first intermediate quantum state, determining a first fine proportion estimated value corresponding to the first sub-quantum state and a second fine proportion estimated value corresponding to the second sub-quantum state so as to realize quantum state preparation. In this way, quantum amplitude amplification is prevented from being directly executed on overall data, and the amplitude amplification times are reduced.

Systems and methods for detecting adversarial artificial intelligence attacks are provided, including identifying a base state having a highest probability amplitude from an output quantum state; performing quantum amplitude amplification on the identified base state based on the output quantum state; for each row in the baseline quantum state lattice matrix, performing quantum amplitude amplification on a base state similar to the identified base state based on the quantum state generated for the row; marking a quantum amplified base state for each of the rows of the line quantum state lattice matrix that exhibits a certain degree of similarity to the quantum amplified identified base state; and determining that the acquired input data includes an artificial intelligence adversarial attack when the reference truth input associated with each of the tagged, quantum amplified base states exhibits a certain degree of dissimilarity from the acquired input data.

The present application discloses a black box quantum state preparation method based on group processing. The method includes: determining a rough proportion estimate of the target state based on a marker bit and an initial quantum state, wherein the initial quantum state is a superposition state obtained by processing an input data vector based on a black box. When the rough proportion estimate is less than a grouping threshold, the initial quantum state is separated and processed according to a first auxiliary bit to determine a first intermediate quantum state, wherein the grouping threshold is determined based on the number of input data vectors, and the first intermediate quantum state includes a target state portion and an orthogonal state portion, the target state portion includes a first sub-quantum state, and the orthogonal state portion includes a second sub-quantum state. Based on the first intermediate quantum state, a first precise proportion estimate corresponding to the first sub-quantum state and a second precise proportion estimate corresponding to the second sub-quantum state are determined to achieve quantum state preparation. In this way, direct quantum amplitude amplification of the entire data is avoided, and the number of amplitude amplifications is reduced.

The application belongs to the technical field of intelligent transportation, and particularly relates to a key scene generation method for an automatic driving vehicle based on quantum reinforcement learning. The method comprises the following steps: step one, constructing a simulation environment; step two, designing a quantum policy network for continuous control; step three, constructing a multi-dimensional sparse reward function; step four, performing interactive training and optimization on the quantum policy network; and step five, outputting a structured test case and performing visual analysis. The application models the generation of key scenes as a dynamic interaction process between an agent and an environment, efficiently explores sparse dangerous behaviors in a high-dimensional state space through a learnable quantum policy network, significantly improves the search efficiency of sparse key scenes in a high-dimensional scene space by introducing a quantum superposition state and a quantum amplitude amplification mechanism, and solves the combination explosion and sparsity problems faced by traditional methods.

The application discloses a quantum state preparation method and device, and the method comprises the following steps: first, obtaining an N-dimensional second data vector to be prepared; then, constructing a quantum state preparation circuit; finally, generating a control pulse of a quantum logic gate in the quantum state preparation circuit, and applying the control pulse to a quantum bit in the quantum state preparation circuit, so that the quantum bit evolves from an initial state to a second quantum state |Ψ e > through a first quantum state |Ψ t >, which reduces the number of quantum state preparation amplitude amplifications, reduces the depth and operation number of the quantum state preparation circuit by constructing the quantum state preparation circuit.

In various embodiments, systems and methods for quantum-based network traffic anomaly detection are disclosed. Embodiments for a network integrity monitor are disclosed that leverage a quantum computing-based network assessment function to evaluate network event data for the purposes of identifying and / or predicting anomalies indicative of network threats. To identify network anomalies, the network assessment function may treat the anomaly identification as a quantum search task by searching the task data using an amplitude amplification quantum search algorithm and / or using quantum machine learning models to infer a threat prediction that may include a single or multiclass classification characterizing the task data. Such classification(s) may be further assessed by the network integrity monitor as the basis to trigger one or more mitigating steps.

The application discloses a kind of method for realizing distributed quantum amplitude amplification and related equipment, method includes: the initial success probability of quantum state is quantified based on target unitary operator, and configuration unitary operator is obtained from target unitary operator based on division parameter;Based on division parameter, Boolean function is split into the subfunction of preset quantity and sequentially distributed to each quantum node;Based on configuration unitary operator, the probability upper limit value of each quantum node is quantified using amplitude estimation;According to probability upper limit value, iteration number is obtained, and according to initial success probability, iteration upper limit is obtained;Based on iteration number, the measurement string corresponding to quantum node is iteratively output by quantum amplitude amplification algorithm, and then target string is verified and determined;Quantum operation result is obtained by summarizing the target string corresponding to all quantum nodes.The application can effectively avoid the dependence of existing distributed quantum computation on quantum communication means such as quantum teleportation, and can be widely applied in the field of data processing.

The invention discloses a safety risk assessment method, system, equipment and medium based on electric power operation, and belongs to the technical field of electric power operation safety, and the method comprises the steps: collecting multi-modal information, such as physiological monitoring data, environmental parameters, behavior videos and the like, of an operator, and employing an empirical cumulative distribution function and a correlation coefficient of a Hellinger-Pearson weighting mechanism to evaluate the safety risk of the operator; effective fusion of high-dimensional heterogeneous data is realized; a personal safety state is mapped into quantum bits, a three-body coupling Hamiltonian model is constructed, dynamic evolution is carried out in combination with a parameterized quantum circuit, and an amplitude amplification mechanism is introduced; a probability-exposure rate-consequence three-dimensional risk matrix is constructed, quantitative analysis of personal injury consequences is realized, and a hierarchical response strategy is formulated in combination with a dynamic threshold value and expert knowledge; the efficiency and the precision of data fusion are improved, the accuracy and the reliability of risk assessment are enhanced, and the accuracy and the sensitivity of an assessment result are improved.

The invention discloses a quantum processor for screening a target lattice vector and a method for solving a shortest vector, and relates to the technical field of quantum computing, the quantum processor comprises an amplitude preparation circuit and an amplitude amplification circuit; the amplitude preparation circuit is used for obtaining a first amplitude of a target lattice vector quantum state and a second amplitude of a non-target lattice vector quantum state, the two-norm of the target lattice vector is smaller than a preset parameter, and the two-norm of the non-target lattice vector is larger than or equal to the preset parameter; the initial state of the amplitude preparation circuit comprises a superposition state of to-be-screened lattice vectors; and the amplitude amplification circuit is used for amplifying the first amplitude of the quantum state of the target lattice vector to be close to 1, so that the target lattice vector can be screened by using a quantum computer.

The invention relates to methods and systems for operating a quantum computing device to sample from an approximate stationary distribution of a Markov chain P having n states, through the construction and execution of a quantum circuit, said method comprising the step of: Initialize an ancilla qubit to the ∣0> state and at least one qubit register in its initial state; Prepare an unitary matrix U, which encodes a discriminant D in block form, allowing polynomial transformations of its singular values or eigenvalues; Apply successively a single qubit gate on the ancilla qubit and the ancilla controlled unitary matrix U on the qubit registers creating a reflection about an approximation of the stationary distribution of the Markov Chain P; and Prepare an approximation of the sought coherent quantum state associated with the stationary distribution, preferably using quantum amplitude amplification

The invention discloses an operating system kernel vulnerability repairing method and device, equipment and a medium, and relates to the field of operating system kernel security, and the method comprises the steps: compiling a verification rule into an intermediate representation through a preset strategy parser, loading the intermediate representation through a microkernel interface to generate a verification strategy, and registering the verification strategy to a target hook function; collecting data and fusing the data to obtain fused data, and extracting the fused data through graph neural network features to obtain initial vulnerability features; obtaining a system state set according to the target hook function, and determining a target high-risk state set by using quantum bits obtained by compressing mapping through an amplitude amplification algorithm; and determining a target vulnerability based on the initial vulnerability feature and the target high-risk state set, analyzing an execution path based on a target vulnerability data flow diagram to construct a target vulnerability chain, generating a target vulnerability report and a target repair suggestion based on the target vulnerability chain, and executing the target repair suggestion on the target vulnerability chain according to the target vulnerability report. And kernel vulnerabilities can be accurately prevented and controlled.

Generalizations of quantum amplitude amplification and amplitude estimation algorithms work with non-boolean oracles (by way of definition, the action of a non-boolean oracle Uφ on an eigenstate |x is to apply a state-dependent phase-shift φ(x); unlike boolean oracles, the eigenvalues exp(iφ(x)) of a non-boolean oracle are not restricted to be ±1). The non-boolean amplitude amplification algorithm preferentially amplifies the amplitudes of the eigenstates based on the value of φ(x). Starting from a given initial superposition state |ψ0, the basis states with lower values of cos(φ) are amplified at the expense of the basis states with higher values of cos(φ). The non-boolean quantum mean estimation algorithm uses quantum phase estimation to estimate the expectation ψ0|Uφ|ψ0 (i.e., the expected value of exp(iφ(x)) for a random x sampled by making a measurement on |ψ0). The quantum mean estimation algorithm offers a quadratic speedup over its counterpart boolean algorithm known in the art.

The invention discloses a Grover algorithm-based dot-lattice chess game implementation method, which comprises the following steps of: performing quantum bit coding on a dot-lattice chess board of a dot-lattice chess game to generate an initial chessboard quantum state vector; then, based on a Grover algorithm, the initial chessboard quantum state vector is processed, and a target chessboard quantum state vector is determined; and finally, according to the target chessboard quantum state vector, determining a target path so as to realize a dot-lattice chess game according to the target path. Therefore, the classical chessboard state is converted into a quantum processable form through quantum bit coding, and the quantum parallelism and the amplitude amplification characteristic of the Grover algorithm are combined, so that the search complexity of the optimal chess moving path is reduced, the search efficiency is greatly improved, and the optimal solution of scores or opponent scores can be quickly found.