17 results about "Q-matrix" patented technology

The present invention discloses a method for minimal-delay fine-grained parallel Givens rotation of ultra-large-scale matrix QR decomposition, relating to the technical field of signal and information processing. The method comprises: an outer loop and an inner loop, wherein the outer loop is responsible for calculating a row vector of a Q matrix and an R matrix, and the inner loop is responsible for the Givens rotation of the two row vectors; the outer loop is decomposed into multiple loop blocks and a multi-level PE unit array is used to realize parallel operation of the outer loop slices, while the parallel calculation of the inner loop slices is realized by a step-by-step decreasing element update module inside the PE unit; the collaborative design of the inner and outer loop slices significantly reduces the idle time of hardware resources and optimizes the generation and transmission of key data, so that the system starts calculation with minimal delay; the method improves the utilization efficiency of hardware accelerators and is suitable for ultra-large-scale matrix decomposition tasks in the fields of signal processing, image processing, and scientific computing.

A system and method for energy-efficient approximate digital JPEG and MJPEG-compression. The system includes a controller unit to control a processing loop for processing image blocks based on a comparison of a current image block to a previous image block. The system includes a quantization unit configured to quantize the frequency domain representation using an approximate quantization process and a quantization (Q) matrix. The quantization unit is configured to: identify, a nearest power of two value for each element of the quantization matrix; generate an updated Q matrix by assigning each element of the quantization matrix with the identified nearest power of two value; and shift each element of the updated Q matrix by a number of bits to generate a quantized frequency domain representation. The number of bits corresponds to the identified nearest power of two for the corresponding element of the updated Q matrix.

The application discloses a spherical decoding detection method and device, electronic equipment and a readable storage medium, and belongs to the technical field of wireless communication. The application obtains a Q matrix and an R matrix by performing orthogonal triangular QR decomposition on a channel response matrix, wherein each column in the Q matrix is arranged according to the size of channel energy; determines a target equalization signal according to the Q matrix and the R matrix; performs an ML path search process according to the target equalization signal to select N branches with minimum metrics as survivor paths; performs ML complement path search on the survivor paths to obtain ML complement paths; takes the survivor paths with minimum metrics as ML paths, and obtains the likelihood ratio information of each bit of each symbol at each layer according to the ML paths and the ML complement paths. The application can reduce the operation complexity of the spherical decoding algorithm while maintaining the decoding detection performance.

The application belongs to the field of artificial intelligence and educational interactive technology, and discloses an intelligent Q matrix generation method, system and terminal fusing question semantic information, which uses BiLSTM+Transformer to obtain mathematical knowledge points in a data set, and uses CRF to correct the annotation type of each word to output the mathematical knowledge points; uses BiGRU+Attention mechanism to extract the relationship of the knowledge points, and outputs the relationship between each knowledge point pair; uses the relationship between the knowledge points to design a redundant knowledge point reduction algorithm, reduces the redundant knowledge points through the reduction algorithm, and generates a Q matrix using the knowledge point set after reduction. The application can achieve a relatively high knowledge point extraction rate by using the Transformer+BiLSTM+CRF method, and can output the relationship between the knowledge point pairs more accurately by using the BiGRU+Attention method.

The present application relates to the field of automation, and discloses an anti-pinch early warning method and apparatus, and an electronic device. The method comprises: determining distance values between sensing points of a sensor module and the sensor module, and constructing a target distance matrix on the basis of the distance values; acquiring an operating state value of a movable end, and determining a target template matrix on the basis of the operating state value; performing differential calculation on the target distance matrix and the target template matrix according to a preset rule to obtain a target difference matrix, and grouping matrix element values in the target difference matrix on the basis of connectivity of the matrix element values in the target difference matrix, so as to obtain Q matrix element value sets; and determining the number of elements in each matrix element value set, and determining an anti-pinch early warning result on the basis of the number of elements. The present application solves the problem in the related art of a high risk of a user being pinched due to the fact that an anti-pinch operation can be started only upon detecting that a movable end of a target device is in contact with a foreign matter.

The application discloses a likelihood-based cognitive diagnosis Q matrix estimation method, belongs to the field of cognitive diagnosis evaluation, and comprises the following steps: obtaining response data of students to questions and a partially defined Q matrix, fitting a G-DINA model to obtain a posterior distribution of a student attribute vector, and calculating a marginal likelihood based on the posterior distribution. Furthermore, the fitting degrees of different q vectors are evaluated through AIC or BIC values, the optimal q vector is selected to gradually fill unknown parts of the Q matrix, and the process is continued until convergence. The method reduces the dependence on experts, improves the accuracy and efficiency of Q matrix estimation, and is suitable for large-scale education evaluation scenarios.

A system and method for energy-efficient approximate digital JPEG and MJPEG-compression. The system includes a controller unit to control a processing loop for processing image blocks based on a comparison of a current image block to a previous image block. The system includes a quantization unit configured to quantize the frequency domain representation using an approximate quantization process and a quantization (Q) matrix. The quantization unit is configured to: identify, a nearest power of two value for each element of the quantization matrix; generate an updated Q matrix by assigning each element of the quantization matrix with the identified nearest power of two value; and shift each element of the updated Q matrix by a number of bits to generate a quantized frequency domain representation. The number of bits corresponds to the identified nearest power of two for the corresponding element of the updated Q matrix.

The invention belongs to the field of knowledge cognition level mining and adaptive teaching systems, and provides a personalized cognition level mining method based on subjective and objective test question collaborative modeling. Obtaining learner-test question interaction data, constructing a knowledge point vector according to the Q matrix, and generating an initial mastering vector based on the student ID; three types of personalized parameters including slip, guess and difficulty are introduced, and the difficulty and the mastery degree are modulated and then spliced with the Q matrix to form a feature sequence; a bidirectional LSTM is adopted to extract time sequence dependence, a knowledge state input update network is combined, and state estimation is dynamically generated; and splicing with a test question Q vector, inputting into a score prediction network, and fusing slip and guess to construct an improved IRT function. Training is supervised by using historical answer records, prediction errors are minimized, an early stop mechanism is introduced, objective question accuracy and subjective question mean square errors are output, multi-dimensional evaluation is realized, and prediction performance and universality are improved.

The application provides a self-adaptive cruise decision method combining Q algorithm and fuzzy state division, comprising the following steps: step 1, establishing a discrete state variable model; step 2, establishing a fuzzy variable model; step 3, initializing the value in the Q matrix, setting the learning rate alpha, the attenuation rate gamma and the reward value r; step 4, calculating the fitness mu i (s) and selecting the state with the maximum fitness mu k (s); step 5, selecting the action to be executed b; step 6, executing the action b to obtain the next time state s' and the reward value r; step 7, calculating the confidence degree omega j and updating the Q value; step 8, discretizing the continuous action; step 9, judging whether the stopping condition of the self-adaptive cruise task is met, if yes, the self-adaptive cruise task is stopped, if not, the step of calculating the fitness is entered again. The method solves the problems that the traditional algorithm has poor effect due to insufficient training samples and unbalanced collection and cannot be implemented due to the limited storage space of the matrix.

A method for multi-objective vector fitting includes the following steps: 1) processing a given dataset to construct a single-objective fitting dataset; 2) determining initial poles and using the initial poles and frequency sampling points to establish a fundamental matrix; 3) separating the real and imaginary parts of the fundamental matrix to construct a real fundamental matrix; 4) performing a simplified QR decomposition on the real fundamental matrix to obtain the first M columns of the complete Q matrix, denoted as Q. s Matrix, and for Q s The matrix is ​​stored; 5) Construct and solve the pole coefficient equation to obtain the pole coefficients; 6) Construct the pole characteristic equation using the pole coefficients and the initial poles to calculate the new poles; 7) Calculate the residues of each matrix element corresponding to the new poles to complete the vector fitting. This invention provides a method for multi-objective vector fitting that significantly reduces the computational complexity and improves computational efficiency.

The present application belongs to the field of artificial intelligence selection in oil exploitation, and provides a method and device for predicting the yield of an oil well, a storage medium and an electronic device, the method comprising: sequentially inputting training set data in a training set into a bidirectional long short-term memory network branch and a time-frequency transformer branch for processing, and outputting a first prediction value and a second prediction value; calculating a Q matrix based on the first prediction value, and calculating a K matrix and a V matrix based on the second prediction value; performing fusion processing on the Q matrix, the K matrix and the V matrix through a multi-head attention mechanism to obtain a fused result; after converting the fused result through a linear layer, obtaining an optimal parameter model through iterative training; and predicting the oil yield of a target oil well through the optimal parameter model to predict the oil yield of the target oil well. Through the prediction method, the oil yield of the target oil well can be accurately predicted.

The invention belongs to the technical field of large model security, and particularly relates to an efficient large model full-process encryption method. Comprising the steps of traversing a large model and performing operation deployment; the GPU calculates an attention score matrix and positions important parameters in a Q matrix and a K matrix; partitioning the Q matrix and the K matrix; the GPU extracts important parameters from each block and constructs a block parameter packet; after the TEE verifies the block parameter packet, OTP encryption is carried out on the important parameters, and encrypted blocks are generated; the GPU restores the encryption parameter to a parameter position corresponding to the original block to form a semi-encryption block; executing parallel linear ciphertext operation on the semi-encrypted blocks to generate ciphertext operation result blocks; the TEE performs block decryption on the ciphertext operation result, then performs nonlinear operation, and performs secondary OTP encryption on the nonlinear operation result to generate a secondary encryption result; and the GPU directly uses the secondary encryption result to carry out linear calculation, and then transmits the result back to the TEE to decrypt until a reasoning result is obtained, so that the security of sensitive parameter encryption is remarkably improved, and the efficiency is remarkably improved.

The invention belongs to the field of artificial intelligence selection in oil exploitation, and provides an oil well yield prediction method and device, a storage medium and electronic equipment, and the method comprises the steps: sequentially inputting training set data in a training set into a bidirectional long-short-term memory network branch and a time-frequency converter branch for processing, outputting a first predicted value and a second predicted value; calculating a Q matrix based on the first predicted value, and calculating a K matrix and a V matrix based on the second predicted value; performing fusion processing on the Q matrix, the K matrix and the V matrix through a multi-head attention mechanism to obtain a fused result; after a fused result is converted through a linear layer, an optimal parameter model is obtained through iterative training; the oil production of the target oil well is predicted through the optimal parameter model, and the oil production of the target oil well is predicted. Through the prediction method, the oil production of the target oil well can be accurately predicted.

The invention discloses a student cognitive diagnosis method, system and device based on knowledge dependence propagation and adaptive Q matrix optimization and a storage medium, and belongs to the technical field of student cognitive diagnosis. The method comprises the following steps: acquiring a student-practice interaction pair comprising a student ID and a practice ID and a knowledge premise graph comprising a plurality of knowledge concepts; inputting the student-practice interaction pair and the knowledge premise graph into a pre-trained cognitive diagnosis model to obtain a student answer result predicted value; wherein the cognitive diagnosis model comprises an embedded vector generation module, a knowledge dependence propagation module, a Q matrix adaptive optimization module and a prediction module. According to the invention, the mastery degree of the students on each knowledge concept can be accurately evaluated, and the correct rate of the answering results of the students can be predicted.

The application provides a cognitive diagnosis method, medium, equipment and product for automatic construction of a Q matrix, relates to the technical field of cognitive diagnosis, and the method comprises the following steps: calculating the correlation score of an exercise and a pre-defined knowledge point, and screening an initial knowledge point set of the exercise by using a large language model; generating a reasoning path for the exercise solution by using the large language model, screening necessary knowledge points on the reasoning path, generating an implicit knowledge point set of the exercise, and constructing a Q matrix by using the initial knowledge point set and the implicit knowledge point set of the exercise; and using the Q matrix for cognitive diagnosis by using a diagnosis model. The application realizes full-automatic construction of the Q matrix in a non-labeled scene, and reduces the dependence of a CDM on data in actual application.

The invention provides a cognitive diagnosis method, medium, equipment and product for automatically constructing a Q matrix, and relates to the technical field of cognitive diagnos.The method comprises the steps that relevance scores of exercises and predefined knowledge points are calculated, and an initial knowledge point set of the exercises is screened through a large language model; generating a reasoning path for exercise answering by utilizing a large language model, screening necessary knowledge points on the reasoning path, generating a hidden knowledge point set of exercises, and constructing a Q matrix by utilizing an initial knowledge point set and the hidden knowledge point set of the exercises; and applying the Q matrix to a diagnosis model for cognitive diagnosis. According to the method, full-automatic construction of the Q matrix in an unlabeled scene is realized, and the dependency of the CDM on data in practical application is reduced.

The invention discloses an attenuation anisotropic medium earthquake forward modeling simulation method based on a fractional order space derivative, electronic equipment, a storage medium and a device. The method comprises the following steps: constructing a quality factor Q matrix; constructing a viscoelastic parameter beta matrix based on the quality factor Q matrix; constructing a stress-strain equation containing a fractional order space derivative of P-SV seismic waves in the two-dimensional xz plane based on the viscoelastic parameter beta matrix; obtaining an attenuation anisotropic medium wave equation containing a fractional order space derivative based on the stress-strain equation in combination with the geometric equation and the motion equation; and carrying out attenuation anisotropic medium seismic forward modeling simulation by utilizing a finite difference method and a pseudo-spectral method based on the wave equation. According to the seismic forward modeling method, the attenuation anisotropic medium wave equation containing the fractional order space derivative is constructed, then seismic simulation is carried out through the pseudo-spectral method and the finite difference method, high-precision seismic forward modeling is achieved, and the precision of seismic forward modeling in a complex medium is improved.