11 results about "Dependable computing" patented technology

The invention belongs to the field of bioinformatics, and relates to a molecular property prediction method based on multi-modal gating and comparative learning, which comprises the technologies of comparative learning, graph neural network, cross-modal alignment, gating attention and the like. Firstly, data standardization and graph construction are carried out, and molecular fingerprint embedding is extracted; secondly, a heterogeneous dual-channel graph coding architecture is adopted, one path captures atom short-range interaction through an attention mechanism, the other path integrates a molecular global structure and long-range dependence, and complementary molecular representation is generated; then, a cross-modal attention mechanism is introduced, bidirectional association of graph and fingerprint features is achieved, and modal weights are adaptively and dynamically distributed through a gating fusion module; and finally, a comparison pre-training strategy is adopted, a molecular graph and fingerprints are utilized to construct a sample pair, and discriminative molecular representation is learned on unlabeled data. According to the method, the accuracy of molecular property prediction is remarkably improved, and an efficient and reliable calculation tool is provided for virtual drug screening and lead compound optimization.

The invention belongs to the technical field of image processing, and particularly relates to a graphite ore grade detection method based on an improved YOLO11 model, C3k2-CAS and Detect-SEAM modules are introduced, the feature extraction capability is enhanced, the expression of different scales of ore textures is optimized, and the perception capability of the model for grade difference is improved, so that the detection precision is remarkably improved; the attention mechanism of the CAS module is adopted to replace the traditional multiplication operation, the calculation complexity and the model parameter quantity are greatly reduced while the detection performance is maintained, and the improved model is more efficient than a baseline model and is more suitable for edge device deployment; a plurality of data enhancement strategies are combined, so that the model can maintain high robustness and stable detection capability in a complex industrial environment; the trained optimization model can be efficiently deployed to graphite ore grade detection intelligent equipment, real-time and accurate industrial field detection is achieved, the mineral separation efficiency is greatly improved, and an efficient and reliable computer vision solution is provided for the intelligent mining industry.

The invention discloses a synthetic lethal prediction method based on non-generality forgetting and low-rank interaction multi-mode adaptive fusion, and relates to the technical field of bioinformatics, computational biology and drug target screening. The method comprises the following steps: firstly, constructing a multi-view gene embedding expression based on a gene graph structure, a hypergraph structure and a knowledge graph; then constructing a difference vector and an interaction vector for any gene pair so as to characterize the difference and potential complementary relationship between the genes; multi-source information fusion is realized by using a deep learning model containing interactive attention, a forgetting gate and a multi-modal fusion mechanism; on this basis, multi-task optimization is carried out through joint classification loss, modal interaction loss and redundancy suppression loss, and optimal discrimination is realized through a dynamic threshold search strategy; and finally, outputting a sorting result of the candidate synthetic lethal gene pairs according to the prediction score. According to the method, the accuracy and generalization ability of synthetic lethal relationship prediction can be effectively improved, and a high-reliability calculation auxiliary tool is provided for anti-cancer drug target screening and gene therapy strategy design.

The invention provides a multi-level sparse matrix data processing method and device, and the method comprises the steps: responding to a matrix loading instruction, and loading original matrix data; in response to the matrix calculation instruction, performing element sparse processing on the original matrix data to determine an intermediate matrix; performing format compression processing on the intermediate matrix to determine compressed format data; performing bit sparse processing on the compressed format data to output quantized element data; performing error compensation processing on the quantized element data to output a target sparse matrix; receiving the target sparse matrix, performing sparse calculation, and outputting a calculation result; and in response to a matrix storage instruction, storing the calculation result. According to the method, sparse processing is carried out on the matrix from the two aspects of elements and bits, meanwhile, the sparse matrix is corrected through an error compensation method, errors caused by sparsification are reduced, sparse matrix calculation is accelerated, the parallel efficiency of calculation is improved, and reliable calculation precision is guaranteed while efficient sparse calculation is achieved.

The application discloses a multi-modal graph fusion driven spatial transcriptome region identification method, and belongs to the field of biomedical informatics and computational biology. The method comprises the following steps: step 1, data acquisition; step 2, data processing; step 3, multi-modal adjacency matrix generation; step 4, differentiable multi-modal graph structure fusion; step 5, variational graph autoencoder training; and step 6, clustering. By constructing a spatial domain identification framework based on multi-modal graph fusion and variational autoencoder, the application significantly improves the accuracy, continuity and biological interpretability of spatial transcriptome clustering analysis, effectively solves the deficiencies of existing methods in multi-modal fusion, graph structure optimization and feature discriminability, and provides a more reliable computing tool for in-depth revealing of tissue spatial heterogeneity in the fields of brain science, tumor microenvironment analysis and other frontier biomedical researches.

A computer-implemented method for providing a system-specific secret to a computing system having multiple computing components is disclosed. The method includes permanently storing a component-specific import key as part of the computing component and storing the component-specific import key in a manufacturing storage system. Upon a request for a system-specific secret for the computing system, the method includes identifying a computing component included in the computing system, obtaining a record related to the identified computing component, determining a system-specific secret protected by a hardware security module, and determining a system-specific auxiliary key. The method further includes encrypting the system-specific auxiliary key with the obtained component-specific import key, thereby creating an auxiliary key bundle, encrypting the system-specific secret, and storing the auxiliary key bundle and the system record in a storage medium of the computing system.

The invention discloses a spatial transcriptome region identification method driven by multi-modal graph fusion, and belongs to the field of biomedical informatics and computational biology. The method comprises the following steps: step 1, data acquisition; 2, data processing; step 3, generating a multi-modal adjacent matrix; 4, carrying out the structure fusion of the differentiable multi-modal graph; step 5, variational graph self-encoder training is carried out; and step 6, clustering. According to the method, the spatial domain recognition framework based on multi-modal graph fusion and the variational autoencoder is constructed, so that the accuracy, continuity and biological interpretability of spatial transcriptome clustering analysis are remarkably improved, and the defects of an existing method in the aspects of multi-modal fusion, graph structure optimization and feature discrimination are effectively overcome; and a more reliable calculation tool is provided for deeply revealing tissue spatial heterogeneity in the field of advanced biomedical research such as brain science, tumor microenvironment analysis and the like.

The invention discloses a redundancy calculation system, a redundancy calculation processing method and related equipment. The system comprises a first computing unit running in a first kernel in a software lock step mode, a second computing unit running in a second kernel in a software lock step mode, an arbitration unit running in a third kernel and a watchdog unit running in a fourth kernel. The second calculation unit monitors the operation condition of the controller according to a technical scheme different from that of the first calculation unit; the first kernel, the second kernel, the third kernel and the fourth kernel are respectively provided with first middleware used for coordinating time synchronization among the kernels, and second middleware used for scheduling the first calculation unit and the second calculation unit in an absolute time scheduling mode. According to the method, environmental interference in the same time space is avoided to a certain extent, a more reliable calculation result can be obtained, the security level is improved through software redundancy, the system failure rate is reduced, and the method has high software execution efficiency.

The invention relates to the technical field of power equipment multi-physical field calculation and intelligent operation and maintenance, and discloses a power equipment electromagnetic loss rapid prediction method, which comprises the following steps: performing physical driving region decomposition on a three-dimensional model of equipment, and constructing a working condition and node-level loss sample set by using a high-fidelity electromagnetic solver; for grid nodes of each region, constructing a prior enhanced feature vector containing geometric and physical interaction items, and training a dual-stage gradient boosting decision tree proxy model; exporting the model as a general format, integrating the model into a heat flow solver, deducing heat source distribution in real time in coupling calculation, and executing online calibration according to prediction uncertainty. According to the method, node-level high-precision prediction of electromagnetic loss is realized through physical prior feature enhancement and double-stage modeling, time-consuming full-order finite element iteration is replaced, electromagnetic heat flow coupling calculation is compressed from hour level to second level, and efficient and reliable calculation support is provided for digital twinning and real-time operation and maintenance of power equipment.

A system and methods are disclosed for end-to-end provable guarantees for safety critical computer platforms. The technical problem addressed by this invention is the need to provide assurance that a complex safety-critical computer platform behaves as intended and is secure, particularly in safety-critical contexts where errors or malicious attacks can have devastating effects. Despite advances in methodologies and toolchains for high-assurance software production, there remains a significant gap between theoretical guarantees and practical implementation at the operation instruction level (e.g., Assembly language instruction, byte-code instruction, digital circuit state-machine instruction etc.). The present invention overcomes this limitation by providing a system for verifying key security properties for safety-critical computer platforms at the operation instruction level without reliance on custom generation pipelines or certified compilers, thereby enabling the practical development of more secure and reliable computer platforms.

The invention discloses a localized environment-based cross-department comprehensive supervision platform and a data processing method. The system architecture comprises an infrastructure layer, a data layer, a support layer, an application layer, a display layer and a security processing module. The infrastructure layer provides safe and reliable computing storage network resources by adopting a creative cloud and an e-government affair external network; the data layer stores and manages supervision data through a supervision basic library and a supervision thematic library; the support layer provides unified tool support and data support capabilities. According to the invention, a localized unified platform is constructed, the problems of cross-department data sharing and business collaboration are solved, interconnection and on-demand sharing, efficient collaboration and accurate differentiated supervision of supervision data are realized, the supervision efficiency and accuracy are improved, and the whole-process data security is guaranteed through multi-level security measures.