14 results about "Hierarchical search" patented technology

This invention discloses a fast beam alignment method based on the simplex method. It achieves rapid beam alignment between communication devices operating on highly mobile platforms through two stages: preprocessing and simplex processing. The preprocessing stage introduces a hierarchical search approach, providing a more reasonable initial solution for the subsequent beam alignment algorithm, effectively reducing the likelihood of the simplex algorithm getting stuck in local optima. Compared to previous beam alignment methods, this significantly reduces the overall algorithm's time complexity and implementation complexity, while maintaining the same order of magnitude of angular error, demonstrating superior performance.

The application provides a multimodal large model compression method and device based on sparse codebook quantization, and relates to the technical field of computer vision, wherein the method comprises the following steps: optimizing a visual encoder to make the weight significance distribution of an arbitrary large-scale visual-language model more concentrated; evaluating the influence of each layer weight of the optimized language model on the output through second-order information, and dynamically allocating the code word quantity of each weight group according to the significance; determining the optimal sparse combination from a large-scale codebook by adopting a two-stage strategy of high-level candidate search and low-level subset refinement; and completing model quantization by combining the combination and sparse coefficients, so as to balance the compression and inference performance. The dynamic code word allocation and hierarchical search method based on sparse coding does not need additional training, can adaptively allocate the optimal sparse code word combination, can keep the model expression ability under extremely low bits, and can improve the quantization compression efficiency and inference performance.

A beamforming design method for intelligent metasurface arrays based on orthogonal time-frequency-space technology includes: aggregating multipath channels under direct illumination into cascaded channels with near-field direction vectors; randomly initializing the phase shift of intelligent metasurface units; establishing an effective channel matrix based on the cascaded channels; then establishing a multi-user and rate objective function; maximizing the objective function through phase shift optimization using a hierarchical search strategy; repeatedly executing the hierarchical search strategy to search and optimize subsequent blocks until the number of intelligent metasurface units in each block reaches a preset value; then performing an exhaustive search in the final block, thus completing one iteration; repeating the iteration until the maximum number of iterations is reached; outputting the final optimal phase shift; and implementing beamforming design based on the final optimal phase shift. This invention can reduce the complexity of beamforming design and ensure reliable transmission of high capacity during high-speed movement.

This application discloses a method and system for spot market bidding decisions in combined heat and power (CHP) power plants, relating to the technical field of CHP power plants. The method includes: acquiring operating data of the CHP unit and market data of the electricity spot market; constructing a heat-power coupling and coal consumption characteristic model based on the operating data and the market data, with rigid heat load supply parameters as the primary constraint; inputting the cleaned operating data and market data, along with energy conversion efficiency, energy consumption characteristics, and feasible domain of CHP output under different operating conditions, into a chaotic hierarchical gray wolf optimization algorithm to initialize the population and implement a hierarchical search strategy; defining a fitness function with profit maximization as the objective, solving for the optimal power output or optimal bid price, and guiding the CHP power plant in making bidding decisions regarding power generation or price. This application aims to solve the problems of unscientific bidding decisions and low profitability of CHP power plants in the electricity spot market.

A method performed by one or more computers. The method comprises receiving a natural language query specifying requirements for a compound; processing the natural language query using a language policy to generate a plurality of representations of candidate compounds that each satisfy at least a subset of the requirements specified in the natural language query. Each representation specifies at least a chemical formula of the corresponding candidate compound. The method further comprises, for each representation in a subset of the representations, using a generative machine learning model conditioned on the representation to generate one or more candidate chemical structures, each candidate chemical structure comprising a respective spatial location for each of the atoms of the corresponding candidate compound; and selecting a chemical structure and corresponding compound from the plurality of candidate chemical structures.

A method and system for cable defect detection and identification based on TMR (Transmission Measuring and Reconstructing) is disclosed. The method includes: receiving high-frequency pulse current signals from a power cable using an array rectangular probe integrating an excitation coil and a TMR sensor array; comparing the signal differences between defective and normal locations to obtain differential signals; and performing optimal spatial reconstruction of the differential signals using phase space reconstruction, symbol entropy, and recursive hierarchical search algorithms to obtain a differential reconstruction space matrix. The HIS (Hyperspective of Influence) distribution curve is then calculated and used as input features for a power cable defect identification model with an adaptive activation function, enabling automatic defect identification. This invention employs phase space reconstruction technology to reconstruct the spatial matrix of the differential signals. By calculating the HIS distribution curve of the differential reconstruction space matrix, the signal distribution differences reflected in the reconstruction matrix caused by fluctuations are eliminated, resulting in an overall 12% improvement in recognition rate compared to traditional methods combining statistical features such as peak voltage with models such as SVM and 1DCNN.

The application discloses a network positioning method and system based on topological correlation analysis, relates to the technical field of communication network fault positioning, and comprises the following steps: acquiring a receiving optical power sequence, a forward error correction count sequence, a logical topological relationship and physical resource organization data of a network branch, calculating a branch drift sequence, and establishing a membership relationship between the network branch and a candidate physical positioning unit; generating a common mode drift sequence based on a nearest common ancestor, and performing common mode processing to calculate a selective synchronization coefficient; further calculating a selective interpretation degree of the candidate physical positioning unit, and combining a hierarchical search graph and a swarm intelligence algorithm to obtain an optimal physical positioning unit; finally, determining an affected branch set and a relative confidence degree, and outputting a positioning result. The application can realize fine positioning of a local physical abnormal position of a PON tree-shaped optical network.

This invention discloses a hierarchical retrieval method, storage medium, and computer device, belonging to the field of document processing technology. It includes: S100, obtaining the hierarchical paragraph parsing results of the input document; S200, retrieving the title paragraphs in the parsing results according to the user's query, generating a set of title paragraphs based on the retrieval results, determining whether the title paragraph set is empty, and if not, outputting a semantically complete paragraph; if so, proceeding to step S300; S300, retrieving the body text paragraphs in the parsing results according to the user's query, and outputting a semantically complete paragraph or summary based on the retrieval results. This invention, by combining the hierarchical parsing results of the document, can improve the completeness of text retrieval results in large-scale model retrieval and enhance document question answering, thereby improving the correctness and completeness of the generated answers in document question answering.

This invention discloses an adaptive GNSS position parameter state update method based on elastic thermal grid search; belonging to the field of GNSS positioning and navigation technology; the method includes: starting a differential relative positioning program, storing fixed solution coordinates and their covariance information in real time; performing a discreteness test on the accumulated historical fixed epochs, and when the fixed solution distribution meets the stability condition, taking the current accumulated number of fixed solutions as the total number of valid samples; using the initial fixed solution of the station as the initial heatmap center, adaptively determining the heatmap search range and step size, constructing a multi-layer thermal grid for layer-by-layer refinement search until the layered search converges, and outputting the coordinate parameters and their covariance corresponding to the center of the grid with the highest heat in the final layer; re-initializing the position parameters with the coordinate parameters and their covariance as initial variables, and continuing to execute differential relative positioning until the observation data solution is completed. This invention suppresses multipath errors and improves the solution accuracy of fixed position parameters.

A method and apparatus for intelligent semiconductor defect analysis, relating to the semiconductor manufacturing field, are disclosed. This invention acquires a design layout, inspection images obtained from an inspection machine, and corresponding re-inspection images obtained from a re-inspection machine, and aligns them using coordinates. Based on the coordinate alignment results, a defect pattern template is generated. The defect pattern template includes defect regions extracted from the re-inspection images, the corresponding positions of the defect regions in the design layout, and the surrounding design environment information of the corresponding positions in the design layout. A hierarchical search is performed using a geometric hash algorithm to locate potential defect locations similar to the defect pattern template in the entire chip's design layout. Precise defect statistics are performed, matching the potential defect locations with the full defect list provided by the inspection machine to determine the total number of specific defect types on the entire wafer. This invention achieves a transformation from sampling analysis to comprehensive intelligent diagnosis, greatly improving the efficiency and accuracy of semiconductor manufacturing process monitoring.

Embodiments of the present application provide a vector search method, a storage device, a storage device cluster and an electronic system. The method is applied to a storage device, the storage device stores a plurality of reference vectors, each reference vector has original data represented in a first data format and a data copy represented in a second data format; the method comprises: determining a target vector to be searched; based on the data copies of the plurality of reference vectors, performing a vector search operation to obtain N candidate reference vectors similar to the target vector; obtaining the original data of each candidate reference vector; based on the original data of the N candidate reference vectors, performing a distance calculation operation to determine the vector distance between each candidate reference vector and the target vector; sorting the N candidate reference vectors according to the vector distance to generate a corresponding vector sequence. By adopting hierarchical search, the search delay caused by full vector cross-device transfer in the traditional architecture can be avoided from the root, and the search efficiency is significantly improved.

This invention relates to the field of neural network architecture search technology, and discloses a differentiable neural architecture search method based on a hierarchical-feedback mechanism and TEG-guided search. The method includes: a hierarchical search stage, where the search network depth gradually increases, and architecture selection is performed using TEG metrics. A differentiable architecture search method is used for neural architecture search. The search process includes multiple search stages, each corresponding to a search network. As the search stages increase, the number of units in the corresponding search network increases, gradually approaching the number of units in the evaluation network used in the evaluation stage. The second step is a feedback search stage: the evaluation network is expanded from the search network of the last search stage. The search network and evaluation network are integrated, and the search network and evaluation network of the last search stage are jointly trained in a cyclical manner. This significantly improves search efficiency while maintaining high accuracy of the search model.

This application discloses a method for assessing small failure probabilities based on prior constraint integration and hierarchical correction sampling, relating to the field of uncertainty quantification technology. The method includes: first, constructing an integrated surrogate model incorporating multiple types of prior constraints and performing residual correction using high-fidelity anchor samples; then, calculating soft failure weights based on the corrected surrogate model, and using this weights to perform hierarchical search of candidate samples to obtain a failure sample set; next, clustering the failure samples to identify multiple failure sample clusters and constructing a multi-scale hybrid proposal distribution; then, performing effective sample size-driven bridging sampling, adaptively controlling the transition step size to obtain a stable initial failure probability estimate; finally, using inverse probability weighted unbiased correction, performing high-fidelity verification of the bridging samples prioritizing false negative risk and correcting the initial estimate, outputting the final failure probability. Under the condition of strictly limited high-fidelity evaluation times, this method achieves high-precision, high-stability, and statistically unbiased estimation of small failure probabilities.

This invention discloses a hierarchical image clustering system and method that integrates geometric and photometric similarity. First, a global feature vector is extracted from each image in the image set, and candidate image pairs are selected based on an approximate nearest neighbor search strategy to limit the scope of fine-grained similarity calculation. Then, a fusion similarity calculation is performed on the candidate image pairs, comprehensively considering geometric structural consistency, photometric appearance consistency, and differences in aligned images to obtain a mixed similarity score. Based on this, a sparse undirected graph structure is constructed, and initial clustering is achieved through graph clustering analysis. For newly added images, an incremental update strategy using hierarchical search is employed for efficient classification. Simultaneously, multi-granular hierarchical structure mining is achieved through recursive clustering analysis of the subgraph structures. This invention balances clustering accuracy and computational efficiency, and is suitable for applications such as large-scale industrial image analysis and defect detection.