14 results about "Data generator" patented technology

The application provides a visual language large model adaptive bias evaluation method for evaluating bias performance of a target model on a specific task, and the method comprises the following steps: S1, obtaining an initial evaluation dataset, wherein the initial evaluation dataset comprises a plurality of evaluation data; S2, fine-tuning a preset large language model using the initial evaluation dataset in a preset fine-tuning manner to obtain an initial data generator; and S3, generating evaluation data based on the initial data generator to perform multiple rounds of bias evaluation on the target model, wherein in each round of bias evaluation, based on all evaluation data corresponding to the last round of bias evaluation, a data generator corresponding to the last round of bias evaluation is iteratively updated according to a preset objective function to obtain a current round data generator, and the current round data generator is used to generate a current round evaluation dataset to perform bias evaluation on the target model.

The invention relates to a system (100) for optimizing the configuration of crash tests and simulations for vehicle development. The system (100) comprises a data preparation module (200) that splits a data set (210) containing crash test and simulation data into a first training data set (210), a validation data set (230), and a test data set (240); a training module (300) with a first AI model (350) for creating configurations of physical crash tests and virtual simulations, wherein the first AI model (350) is trained with the first training data set (220); and a data value module (400) that calculates data values ​​(φ(i)) for the data points (i) of the crash test and simulation data of the training data set (220) with respect to the validation data set (230).a model optimization module (500) comprising a second AI model (550) and a training data generator (570), wherein the training data generator (570) creates a second training data set (580) which is used to train the second AI model (550) to predict the relationship between input parameters (p) and the computed data values ​​(φ(i)); a parameter module (700) which uses the trained second AI model (550) to find crash test parameters with a maximum data value (φ(i)).

The application relates to the technical field of image processing, and discloses a light-weight single-pixel imaging method based on frequency domain-space domain modulation attention, which comprises a single-pixel image data generator and a calculation reconstruction generator; an initial inversion estimation is obtained through inverse mapping of a measurement matrix; then, the initial inversion estimation is input into a dual-domain collaborative attention deep reconstruction network; the network is composed of a plurality of dual-domain collaborative attention module cascades connected in series, a multi-head spectrum calibration module in the module realizes long-distance information reconstruction through dual-domain feature information fusion interaction; a multi-scale-depth dynamic convolution fusion module extracts features in a space domain, accurately processes high-frequency edges, and significantly enhances the reconstruction capability of texture details; and the features are corrected in combination with a residual error guiding update module. Through the dual-domain collaborative mechanism, the reconstruction precision and the detail recovery capability of single-pixel imaging are effectively improved while the network parameter quantity and the calculation complexity are greatly reduced, and the application is suitable for edge deployment.

The present application belongs to the technical field of oil and gas exploration seismic data interpretation. The present application discloses a method and device for identifying faults based on differential cepstrum. The method performs differential cepstrum processing on seismic data of a target layer section, extracts first-order differential cepstrum coefficients of the differential cepstrum, sets a processing factor to calculate eigenvalues of a covariance matrix of the first-order coefficient data body of the differential cepstrum, and generates a high-resolution fault information data body. The differential cepstrum operation adopted by the method strengthens the weak information singularity feature, can identify small faults, and is beneficial to improving the accuracy and precision of fault identification of seismic data. The device for identifying faults based on differential cepstrum provided by the present application comprises a Fourier transform spectrum generator, a logarithmic spectrum generator, a Fourier inverse transform processor, a differential cepstrum first-order coefficient data body generator, and a fault data generator. The device realizes the method for identifying faults based on differential cepstrum.

Semiconductor devices, controllers, and their operating methods are disclosed. In an embodiment, a controller includes an internal command generator configured to generate overwrite operation setting information in response to a data permanent erase request to perform an overwrite operation on a data storage area corresponding to the data permanent erase request; an overwrite data generator configured to generate a plurality of unit addresses corresponding to the data storage area based on the overwrite operation setting information, and to generate a plurality of pieces of overwrite data to be written to the data storage area corresponding to the plurality of unit addresses, respectively; and a storage area controller configured to control the data storage area to store the plurality of pieces of overwrite data corresponding to the plurality of unit addresses, respectively.

A semiconductor device, a controller, and an operating method thereof are disclosed. In an embodiment, a controller includes an internal command generator configured to generate rewrite operation setting information to perform a rewrite operation on a data storage area corresponding to a data permanent erase request in response to the data permanent erase request, a rewrite data generator configured to generate a plurality of cell addresses corresponding to the data storage area based on the rewrite operation setting information, and generate a plurality of pieces of rewrite data to be respectively written to the data storage area corresponding to the plurality of cell addresses, and a storage area controller configured to control the data storage area to store the plurality of pieces of rewrite data corresponding to the plurality of cell addresses, respectively.

This invention relates to a knowledge rewriting method for complex knowledge graph question-answering tasks, specifically as follows: Step 1, using GLM-4 as a data generator to construct a dataset for supervised training; Step 2, based on the constructed dataset, supervising training an open-source large model to enable it to initially master knowledge rewriting capabilities; Step 3, sampling multiple outputs of the same knowledge rewriting task from the supervised-trained large model as candidate knowledge representations; Step 4, using these candidate knowledge representations as context for the question-answering task to obtain the answers corresponding to the questions, constructing a preference dataset; Step 5, rewriting the preference knowledge representations using GLM-4 to improve the quality and diversity of the dataset; Step 6, using the preference dataset and the DPO algorithm to fine-tune the open-source large model to align it with the preferences of the question-answering model. This significantly improves the performance of the large model in handling complex knowledge graph question-answering tasks.

This invention provides a federated learning method using synonymous data, comprising: a coordinating device sending a general model to each client device; each client device executing a training procedure, including: an encoder encoding private data into a summary; training a client model based on the private data, the summary, and the general model; and sending the summary and client parameters of the client model to the coordinating device; the coordinating device identifying absent client devices among the client devices; generating synonymous data using a synonymous data generator based on the summary corresponding to the absent client device; training an alternative model based on the synonymous data and the summary corresponding to the absent client device; and performing an aggregation operation based on the alternative model parameters and the client parameters of each client device other than the absent client device to generate update parameters to update the general model; this invention addresses the problem of client departure by synthesizing representative client data in a coordinator.

A defect detection module may include a data collector configured to collect reflected signal data of an amplitude-mode ultrasonic wave for a test subject, a data preprocessor configured to generate merged reflected signal data by extracting a portion of the reflected signal data, assigning numbers to the extracted portion of the reflected signal data, normalizing amplitudes of the extracted portion of the reflected signal data, and then merging the extracted portion of the reflected signal data, a training data generator configured to generate a training dataset by detecting a position of a defect within the test subject by performing principal component analysis on the merged reflected signal data, and classifying the merged reflected signal data into defect-including data and defect-free data, and a defect detector configured to detect a defect in the test subject through the training dataset by using a deep learning model.

The present application relates to the technical field of data privacy protection, and particularly relates to a large model guided biological data privacy protection method and system, which converts time series signals in biological data into medical semantic description texts by using a large model, so that each client generates robust biological data samples by using a data generator locally, and performs local model training based on the robust biological data samples and local biological data samples; each client uploads model gradients of the local model training to a server, the server performs double verification on the model gradients of each client and aggregates the model gradients by using an adaptive aggregation strategy, updates a global model, distributes the updated global model back to each client for the next round of federated training, and stops until convergence. The present application realizes privacy-utility collaborative optimization through semantic data reconstruction, and provides an auditable and landable new paradigm for federated learning in a medical Internet of Things scenario, which takes into account high privacy compliance and clinical usability.

The application relates to an FPGA-based EXFAT file system storage system and method, and belongs to the technical field of data storage. The system comprises an FPGA master control board card serving as a master controller and an NVME SSD serving as a storage medium; the FPGA master control board card and the NVME SSD are inserted into a PCIE SWITCH adapter plate; wherein the FPGA master control board card is internally integrated with a MicroBlaze system, an NVME HA, an XDMA and a plurality of hardware function IPs; wherein the plurality of hardware function IPs comprise AXI4 protocol packaging IPs, DMA setting IPs, SQ generator IPs, file creation IPs, FAT table marking refreshing IPs, cluster bitmap marking refreshing IPs, SQ chain table IPs, command issuing IPs and data generator IPs. The system realizes a high-speed, low-delay, low-power-consumption and high-reliability EXFAT file system storage system.

The present invention provides an electric motor evaluation index display device that can evaluate the margin of output of an electric motor using a simple structure and method. [Solution] The system includes an output measuring instrument 2 for measuring the output of the electric motor 210 or the output of the vehicle 200 equipped with the electric motor 210, and / or a temperature measuring instrument 1 for measuring the surface temperature of the electric motor 210, and a margin data generator 3a for generating and outputting margin determination data for determining the magnitude of the margin of a predetermined output to be evaluated in the electric motor 210. After driving the electric motor 210 for a predetermined time in an output range including the output to be evaluated, the margin data generator 3a generates data on the decrease in the output of the electric motor 210 or the output of the vehicle 200 equipped with the electric motor 210 as margin determination data if the output measuring instrument 2 is provided, and generates data on the surface temperature of the electric motor 210 as margin determination data if the temperature measuring instrument 1 is provided.