2410 results about "Data encoding" patented technology

A method is provided for enhanced data routing in neural networks using Self-Organizing Maps (SOM) integrated with Autoencoder-GAN. The method comprises training an autoencoder to encode input data into a latent space representation; applying a Self-Organizing Map (SOM) to organize the latent space representation into a topological map; refining the latent space representation using a Generative Adversarial Network (GAN), wherein the generator generates enhanced latent space representations and the discriminator evaluates their quality; using the refined latent space representations to update the SOM topology dynamically; generating routing coefficients based on the updated SOM topology to guide data routing in a capsule network; and dynamically adjusting routing within the capsule network using the generated routing coefficients to enhance performance based on the refined latent representations.

A system and method for temporal acceleration encoding in Lorentzian latent space enables real-time event forecasting within navigable spatiotemporal media. The system encodes media data into compact Lorentzian latent patches using variational autoencoders and organizes them within a multi-dimensional hyperspace spanning spatial, temporal, orientation, scale, and spectral coordinates. Temporal acceleration encoding computes velocity and acceleration vectors along geodesic trajectories, extracting event signatures through multi-scale aggregation over sliding windows. An acceleration-indexed memory stores dynamic descriptors with composite keys comprising hyperspace coordinates and motion characteristics. Event forecasting retrieves similar historical patterns and conditions a forecast head to produce event probabilities and time-to-event estimates with uncertainty calibration. The system streams forecast metadata to edge devices for real-time prediction and adaptive navigation, supporting applications in surveillance, autonomous systems, predictive media exploration, and anomaly detection where both temporal forecasting and multidimensional navigation capabilities are essential.

The invention provides an underwater target detection method based on multi-modal features and domain adaptation. The method comprises the following steps: S11, acquiring a sonar image, an optical image and environmental data; s12, extracting a sonar feature and an optical feature, encoding the environment data into an environment channel weight, and dynamically adjusting a fusion proportion of the sonar feature and the optical feature through the environment channel weight to obtain a fusion feature; s13, performing spatial attention calculation on the sonar features to obtain a spatial weight map, enhancing the optical features by using the spatial weight map, and performing forced alignment with the sonar features at the target edge; and S14, decoupling the fusion feature into a synthetic domain feature, decoupling the fusion feature and the environment data into a real domain feature, and gradually aligning the synthetic domain feature and the real domain feature through asymptotic domain alignment to complete construction of a target detection model. According to the invention, multi-modal data acquisition, dynamic feature fusion and decoupling and embedded real-time detection are combined, so that the precision of underwater target monitoring is remarkably improved.

A neural network system is provided. The system includes an autoencoder configured to encode input data into a latent space representation; a generator neural network configured to receive a noise vector and the latent space representation and output a set of routing coefficients; a discriminator neural network configured to evaluate the effectiveness of the routing coefficients by measuring the performance of a capsule network utilizing said routing coefficients; and a capsule network comprising a first capsule layer and a second capsule layer, wherein the routing coefficients are used to dynamically route outputs from the first capsule layer to the second capsule layer.

Provided in the present invention are a method for training a natural language processing model, and a method for generating subsequent text of a dialogue. The method for training a natural language processing model comprises: acquiring multiple types of heterogeneous sample data and a pre-constructed natural language processing model, wherein the heterogeneous sample data includes structured data, unstructured data, a knowledge graph and expert experience data; performing data encoding fusion on the structured data, the unstructured data and the knowledge graph, so as to obtain encoding fused data; and on the basis of the encoding fused data and the expert experience data, training the pre-constructed natural language processing model, so as to obtain a trained natural language processing model. The present invention enables a knowledge graph to be placed in a model, not as an independent retrieval corpus, but as a method for knowledge enhancement, thereby improving the efficiency of a natural language processing model.

The invention provides a private weight self-adaptive heterogeneous data federated cooperative training method and system in the technical field of federated learning and privacy computing, and the method comprises the steps: S1, enabling each client to carry out the differential privacy operation on a local data set based on a private weight, and obtaining a desensitized data set, encoding the desensitized data set through a heterogeneous data encoding model; s2, performing semantic alignment on each coding vector through a contrast learning model to obtain an aligned vector set; s3, training a local model through the alignment vector set, generating a local gradient, extracting local model parameters, and uploading the privacy weight, the local gradient and local difference parameters to a server; and S4, the server trains the global model based on the local difference parameter and the global gradient, extracts the global model parameter and issues the global model parameter to each client for training. The method has the advantages that the compatibility, the flexibility and the efficiency of heterogeneous data federation cooperative training are greatly improved.

The invention discloses a multi-modal data processing method and system, computer equipment and a readable storage medium, which can realize deep association and complementarity mining of multi-modal information and improve the accuracy and robustness of multi-modal understanding. The method comprises the following steps: an environment sensing module adjusts an environment sensing strategy according to feedback information transmitted by a self-adaptive decision module, and acquires multi-modal data according to the environment sensing strategy; the multi-modal encoding module encodes the multi-modal data into multi-modal feature vectors of the same dimension; a cross-modal fusion module fuses the multi-modal feature vectors to obtain fusion features; the self-adaptive decision-making module selects a decision-making network matched with the task type from a predefined network library according to the task type of the current decision-making task, inputs the fusion features into the decision-making network, and generates feedback information according to the decision-making process of the decision-making network; and the meta-learning controller evaluates the system performance of the current multi-modal data processing system and adjusts system parameters according to an evaluation result.

The invention relates to the technical field of surgical robot motion control, and discloses a surgical robot motion planning method based on a diffusion model, and the method comprises the steps: obtaining obstacle point cloud data, an initial posture and a target posture in a surgical environment; encoding the obstacle point cloud data into a potential space, and combining the encoded information with the initial attitude and the target attitude into a condition code; a Transform structure of an encoder is adopted to replace a U-Net structure in a traditional diffusion model, and the diffusion model is trained through forward diffusion and reverse denoising processes; in training, using a comprehensive loss function to optimize model performance, combining configuration space loss, geometric task space loss and collision loss, and introducing physical constraints to ensure that the trajectory conforms to kinematic characteristics and avoid collision; and in the operation task, generating a motion track based on the trained diffusion model. According to the invention, the motion planning efficiency and safety of the surgical robot in a complex environment can be improved.

The invention provides a multi-modal automatic driving scene generation method based on autoregressive closed-loop prediction. The method comprises the following steps: encoding an automatic driving scene video, scene understanding prediction questions and answers and trajectory planning data into a unified multi-modal discrete code by using a pre-trained discrete encoder; multi-modal discrete codes are integrated into a coding sequence, a mask strategy is adopted to cover trajectory codes, original codes without masks serve as a supervision target, a model is generated through autoregressive normal form training, and fine tuning is carried out on scene understanding, scene prediction and trajectory planning data sets; during reasoning, an initial scene image collected by a vehicle camera and a user prompt word are input, the model generates a future scene discrete coding sequence according to initial information, and closed-loop prediction is realized through mask track information; the generated codes are decoded into scene images, text questions and answers, and vehicle trajectories using a symmetric decoder. According to the invention, the problems of poor universality and low efficiency of the existing end-to-end automatic driving world model can be solved.

The invention belongs to the technical field of intelligent traffic system and artificial intelligence crossing, and particularly relates to a traffic multi-agent simulation decision-making method and system based on a large language model. Road network state data are coded into a three-dimensional feature matrix containing channel dimensions, time dimensions and space dimensions, and joint representation of numerical road network data and text event reports is achieved through a hybrid embedding model. The decision-making layer comprises a dynamic Prompt generator which generates a candidate scheme set based on a four-layer progressive prompt structure; the Monte Carlo tree search multi-objective optimization is executed in cooperation with the decision core module; and the execution layer comprises a cross-language communication bridging device which adopts a gRPC bidirectional stream communication protocol and a Protobuf data serialization scheme to realize millisecond-level data interaction between services and support a hot plug mechanism of a strategy injection interface. According to the invention, dynamic optimization of urban traffic resource allocation and breakthrough improvement of simulation deduction efficiency are realized.

The invention provides a communication index prediction method based on a multi-modal large model and related equipment, relates to the technical field of data processing, and can obtain multi-dimensional time sequence data, geographic space data and corresponding text description data in a user communication process; performing time sequence feature extraction processing on the multi-dimensional time sequence data to obtain a time sequence feature vector, and performing spatial feature extraction processing on the geographic spatial data to obtain a spatial feature vector; encoding the text description data based on a large language model word embedding layer to obtain a semantic feature vector; performing cross-modal attention fusion on the three types of feature vectors based on a text prototype to generate joint feature representation; and the joint feature representation is input into the large language model to generate communication index prediction information, so that the coupling relationship among the multi-modal indexes can be effectively modeled, the prediction precision is improved, and the effectiveness of a network optimization decision is ensured.

The invention discloses a joint denoising method for robot visual motion prediction, and the method comprises the steps: constructing a unified generative model through fusing an image and a depth map collected by a depth camera, motion data collected by CAN line communication of a Piper mechanical arm, and a tactile image collected by a Gelsight Mini tactile sensor; the method comprises two steps of data acquisition and input coding, and joint denoising and generation: firstly, multi-modal data are coded into low-dimensional potential representation, and then future images, depth maps, tactile data and robot actions are cooperatively predicted through a joint denoising framework based on Transform. A mask self-attention mechanism is innovatively introduced, information interaction between modes is dynamically adjusted, action generation is guided through tactile feedback, and the force control precision is improved. The model adopts a de-noising diffusion probability loss function to jointly optimize multi-modal prediction, so that the output consistency is ensured. According to the method, the robustness and the accuracy of flexible operation of the robot are remarkably improved.

The invention discloses a product collaborative design management method and system based on data driving, and relates to the technical field of product design management, and the method comprises the steps: constructing a double-flow heterogeneous quantitative collection network; establishing an incidence matrix and calculating dynamic coupling strength among the parameters; designing a recursive deep belief network based on the incidence matrix and the dynamic coupling strength, encoding product structure parameters and design process data into a probability graph model, and constructing a design knowledge base; dynamically distributing the design rules in the design knowledge base by adopting a swarm intelligent optimization algorithm, generating a collaborative decision-making unit, and establishing a constraint propagation link; and generating a multi-target collaborative optimization scheme group, and screening an optimal scheme group based on Pareto frontier. According to the invention, omnibearing acquisition of product structure parameters and design flow data is realized through the double-flow heterogeneous quantitative acquisition network, and the data integrity is improved.

The invention provides an L1-L4 lumbar vertebra segment-oriented multi-feature fusion diagnosis system and method, and the system comprises an image preprocessing module which is used for receiving a lumbar vertebra CT image sequence of a patient; a centrum anatomy partition module; the multi-dimensional image feature extraction module is used for extracting four types of quantitative features from each sub-region; the clinical multi-modal data coding module is used for independently acquiring and processing three types of clinical data: a multi-modal graph attention fusion network; and the segment-level diagnosis output module outputs diagnosis results of three levels. Through a parallel processing architecture and an optimized feature extraction algorithm, the whole diagnosis process only needs 45 seconds from data input to report generation, time is saved compared with manual film reading, and the consistency of diagnosis results is remarkably improved.

The invention relates to the technical field of computer system structures and storage, in particular to a non-volatile memory quick recovery method based on metadata priority and on-demand loading. The method comprises the following steps: in response to a system fault signal detected by a voltage monitoring unit, freezing a processor context and traversing a page table structure to extract system configuration information; writing the system configuration information and business data codes in the volatile memory into a nonvolatile medium to generate a persistent state mirror image; analyzing the persistent state mirror image, extracting address conversion metadata, and reconstructing a mapping relation from a virtual address to a physical page frame in a volatile memory; and generating an address mapping table, wherein the physical page frame pointed by the address mapping table is set to be in an existing state but is not associated with the effective service data. According to the method, decoupling of the control flow and the data flow is realized by constructing a virtual ready state, and quick starting of the system and immediate response of key services are realized on the premise of not depending on the total capacity of a memory.

The invention relates to the technical field of urban earthquake disaster simulation, and provides a visualization system and method for urban building group earthquake disaster simulation, and the system comprises an offline data preprocessing and block organization module and a block dynamic data management and scheduling module. The GPU acceleration rendering module is used for mapping a global floor index to a vertex to endow the vertex with a unique identifier, and finally submitting the combined model to a GPU, and cooperating with a displacement texture and a user-defined shader to realize real-time rendering of a building form; the GPU parallel displacement picture moving module is deeply coupled to the GPU rendering pipeline and is used for resolving the vertex displacement and the color value in real time and injecting the vertex displacement and the color value into the rendering pipeline; and the efficient interaction module dynamically updates label contents and coordinates to realize three-dimensional interaction. According to the method, massive time sequence displacement data are coded into textures, and parallel calculation is performed by using the GPU vertex shader, so that real-time and smooth animation simulation of earthquake responses of all buildings is realized, and the performance bottleneck of a traditional CPU calculation mode is solved.

The invention relates to the field of artificial intelligence and financial science and technology, and discloses an asset assessment method based on a multi-modal large model, comprising the following steps: preprocessing image data, text data and structured data of a to-be-assessed enterprise to generate standardized multi-modal input data; based on the standardized multi-modal input data, respectively processing image data, text data and structured data by using an image encoder, a text encoder and a structured data encoder, and extracting a high-dimensional feature vector of each modal; mapping the obtained high-dimensional feature vector of each mode to the same public embedding space by constructing a mapper network; an optimal transmission method is adopted to measure the embedding distribution difference between each mode, and the difference is minimized. According to the method, the multi-modal encoder, the unified mapper and the optimal transmission alignment mechanism are constructed, so that high-consistency fusion and expression of the multi-modal features in the shared space are realized.

The invention discloses a bearing life prediction method and system based on dynamic knowledge embedding, and the method comprises the following steps: S1, encoding bearing field data, expert experience and monitoring data into a structured triple, building a dynamic knowledge graph frame, and designing a sliding window confidence mechanism to achieve the online updating of a graph node relation; s2, extracting knowledge embedding vectors by using a relational graph convolutional network, and extracting vibration signal features by using a hierarchical convolutional network; s3, mapping the knowledge embedding vector and the vibration characteristics to a unified semantic space through a linear projection layer; s4, constructing a Transform encoder based on multi-head self-attention, and establishing a dynamic correlation model between vibration characteristics and knowledge embedding; s5, designing a full-connection network output life prediction result and feeding back the optimized knowledge graph; and S6, adaptively adjusting the size of the sliding window based on the change rate of the working condition, dynamically balancing the contribution weight of new and old knowledge in combination with a gating mechanism, and ensuring the adaptability of the model to the complex working condition.

The invention discloses an intelligent power grid optimization scheduling method based on digital twinning, and relates to the technical field of power grid optimization scheduling, and the method comprises the steps: building a dynamic power grid mirror image model according to a standardized power grid data matrix, and generating an evolution trajectory prediction report through a quantum magnetic coupling mechanism; establishing a variable association network based on the evolution trajectory prediction report, and generating an optimization scheduling instruction set in combination with the standardized data matrix; executing an optimization scheduling instruction set to obtain magnetic potential gradient distribution, and generating a topology reconstruction scheme by identifying a magnetic potential change trend; and inputting the topology reconstruction scheme into the dynamic power grid mirror image model to generate an electromagnetic field evolution trajectory, and generating a safety scheduling report in combination with magnetic potential energy gradient distribution. According to the method, the power grid magnetic field data is coded into an evolution trajectory prediction report through a quantum-magnetic coupling mechanism, and high-precision risk pre-judgment is realized; a topology reconstruction scheme is dynamically generated according to the electromagnetic energy gathering trend, and the power grid optimization scheduling efficiency is improved.

The invention provides a retimer and an electronic device, when the retimer enters an error code detection mode, the retimer is switched to a loopback mode, so that the output end and the input end of a second physical layer module are conducted; when a first-order test is carried out, a code stream generation module generates a pseudo-random digital stream according to first source data information in a first source register, and transmits the pseudo-random digital stream to a first data coding module; and the first code stream detection module verifies an output code stream of the second data decoding module according to the first source data information to confirm whether an error code exists or not. By generating and transmitting the pseudo-random digital stream, whether an error code occurs in a link transmission process among the first data coding module, the second physical layer module and the second data decoding module in the retimer is determined, so that when the error code exists, a module which may have a fault can be determined, and error code fault positioning is realized. And a worker can repair the device conveniently.

The invention discloses a non-contact multi-modal decoupling emotion recognition method and device in a dialogue scene. The method comprises the following steps: acquiring original data of multiple modals in the dialogue scene; encoding the original data into original features by using a mode-dedicated encoder; projecting the original features by using a shared feature projector to obtain projection features, and performing weighted fusion to obtain shared features; extracting exclusive features from the original features by using a modal-specific expert network, and carrying out weighted fusion on the exclusive features to obtain private features; fusing the shared features and the private features through a cross attention fusion module to obtain multi-modal fusion features; and classifying the multi-modal fusion features by using a first classifier to obtain an emotion recognition result. According to the method, the key problems of high modal feature heterogeneity, inconsistent modal information, unbalanced modal, missing and the like in the field of multi-modal emotion recognition are solved, and the performance and robustness of emotion recognition in a dialogue scene are improved.

The invention discloses a quantum fuzzy neural network adaptive to high-dimensional input and a classification method, and relates to the field of quantum calculation and fuzzy neural networks and the field of computer vision. The network input layer receives high-dimensional data, amplitude coding, forward and reverse enhanced chain entanglement layer, parameterized quantum transformation and fuzzy set mapping are carried out through a quantum fuzzy feature extraction module, and dynamic dimension fuzzy features are output; high-dimensional neural features are extracted through a DNN feature extraction module to adapt to quantum fuzzy feature dimensions; dynamically distributing the weights of the quantum fuzzy features and the classic neural features through an adaptive feature fusion module; and carrying out Softmax classification on the fusion features through a classifier, and outputting a category probability. According to the method, the high-dimensional data coding efficiency can be effectively improved, the complex fuzzy logic relation learning capability of the quantum part and the quantum state correlation stability are enhanced, the uncertainty of the data is represented, and accurate classification of high-dimensional uncertainty images is realized while noise interference is reduced.

The invention provides an intelligent gateway data processing method for industrial multi-protocol adaptive conversion, which aims at the current situation that various heterogeneous communication protocols exist in an industrial site and is based on a complete process of protocol feature vector extraction, machine learning automatic identification, semantic level analysis and mapping, dynamic rule configuration and protocol frame reconstruction. By constructing a protocol feature model library, semantic contents of original data frames are automatically identified and analyzed, and accurate equivalent conversion of multi-protocol data is further realized in combination with a protocol conversion rule and a unified semantic model. The reconstruction function frame realizes frame structure assembly, data coding and check calculation, and ensures that an output frame completely conforms to a target protocol specification. The method has high compatibility, high expansibility and good real-time performance, and the industrial protocol intercommunication efficiency and the automation degree are remarkably improved.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

A computing system is disclosed for generating fulfillment-ready coordination sessions based on tokenized item data and user-submitted participation parameters. The system ingests structured, semi-structured, or unstructured item data from merchant sources and applies schema-aligned transformation logic to normalize the data into structured item representations. The normalized records are tokenized into machine-readable item tokens that encode fulfillment constraints and canonical attributes. The system receives user item selections and associated participation parameters, encodes the item tokens and participation data into structured vector embeddings, and applies compatibility scoring logic using vector comparison and rule-based threshold evaluation. Compatibility scores are evaluated against session eligibility constraints derived from the item tokens. When eligibility conditions are met, the system generates a coordination session payload comprising match outcomes, proportional pricing, and fulfillment metadata, and issues orchestration instructions to external fulfillment systems for group-based delivery or preparation execution.

The invention belongs to the technical field of industrial automation, and discloses an optimization application method and device for an OPC UA protocol in the industrial automation field, and the method comprises the steps: obtaining the multi-dimensional context information of an industrial site, analyzing the future communication demands and behavior trends of a target OPC UA node through a preset prediction model, obtaining predictive behavior features, and carrying out the optimization application of the OPC UA protocol. And according to the predictive behavior characteristics and the current context information, an OPC UA communication optimization strategy is dynamically generated and adjusted, the OPC UA communication optimization strategy comprises a data sampling period, a release interval, a message queue size, a data coding mode and a security strategy level, and the optimization strategy is applied to a communication session between an OPC UA server and a client. According to the method, OPC UA communication parameters are dynamically optimized through intelligent perception and predictive analysis, the data transmission efficiency can be remarkably improved, the system resource consumption is reduced, the real-time performance and reliability of communication are enhanced, and the method is particularly suitable for a complex and dynamic industrial automation environment.

The invention provides a vehicle track prediction method and system, computer equipment and a medium, and the method comprises the steps: obtaining a track data set, extracting vehicle lane changing data points from the track data set, carrying out the prediction of a driving behavior intention according to the vehicle lane changing data points and a vehicle track data coding result, and obtaining a driving behavior intention prediction result; and correcting the vehicle trajectory data coding result, and performing vehicle trajectory prediction based on the corrected vehicle trajectory data coding result and the driving behavior intention prediction result. According to the method, the vehicle trajectory prediction is performed based on the vehicle trajectory data coding result and the driving behavior intention prediction result, the historical vehicle trajectory data can be combined, and the driving behavior intention of the driver is fully considered, so that the precision of vehicle trajectory prediction is improved, and the method can adapt to more complex traffic scenes. The method can respond to the change of the traffic environment in real time, can adjust the prediction result in time, and guarantees the safety and smoothness of vehicle driving.

The invention relates to the field of medical data analysis, in particular to an interventional therapy patient perioperative period pain data analysis method and system. An interventional therapy patient perioperative period pain data analysis system comprises a data acquisition module, a weight acquisition module, a pain prediction module and a pain monitoring module. According to the method, a standardized multi-source data coding system is established, pre-operative static characteristics and intra-operative dynamic time sequence signals are subjected to nonlinear weight calculation based on mutual information coefficients, complex threshold correlation and interaction effects among parameters are captured, and then a parallel architecture of an LSTM network and a static weighting layer is adopted; deep correlation information of a time sequence dependence mode generated by intraoperative operation and basic background parameters is extracted, the recognition ability of the model for high-risk patients is improved, and high-precision prediction of the pain intensity and type in the perioperative period is achieved.

Techniques for encoding metadata representing a policy into a QUIC connection ID are described herein. A metadata-aware network including one or more enforcement nodes, a policy engine, and / or a connection datastore may be utilized to enforce a policy and route communications on a QUIC connection. The policy engine may be configured to encode metadata representing one or more network policies into a QUIC source connection ID (SCID) and / or may store a mapping between the SCID and a corresponding destination connection ID (DCID) in the connection datastore. The policy engine may communicate with a QUIC application server and / or one or more QUIC proxy nodes to encode the SCID into a QUIC packet. The enforcement nodes may access the metadata and enforce the policies via a connection ID included in a QUIC header of a QUIC packet or by performing a lookup in the connection datastore using the connection ID.