52 results about "Graph state" patented technology

The invention relates to the technical field of respiratory system risk prediction, and provides a respiratory system risk prediction method and system based on a graph neural network, and the method comprises the steps: collecting the multi-modal medical data of a patient, and constructing a multilayer heterogeneous graph based on the multi-modal medical data; constructing a weighted adjacency matrix and a node feature vector through the multi-layer heterogeneous graph; matrix product operation and convolution operation are carried out based on the weighted adjacent matrix and the node feature vector, splicing combination with historical moment state information is carried out, graph state representation is obtained, weighted aggregation of time dimensions is carried out, and time sequence attention features are obtained; performing coding processing based on the clinical examination data to obtain multi-modal fusion features; and inputting the multi-modal fusion features into a risk classifier for classification calculation to obtain a respiratory system risk level prediction result, generating a risk assessment report, and outputting respiratory risk early warning information. The accuracy and clinical practicability of respiratory system risk prediction are improved.

The invention relates to the technical field of multi-agent cooperative control, and discloses a multi-agent reinforcement learning formation adaptive method fusing a graph attention mechanism. According to the method, a dynamic graph state space is constructed, and a speed self-adaptive communication mechanism and a multi-target return function are combined, so that the problem of insufficient generalization ability during intelligent agent number change or topological structure dynamic adjustment is solved. The scheme comprises the following steps: establishing a virtual leader-following formation model and a motion dynamics model; designing an adaptive communication radius formula based on speed change; constructing a graph state space, and dynamically aggregating neighbor information by using a multi-head attention mechanism; designing a strategy network and a value evaluation network with residual connection; and a centralized training distributed execution framework is adopted to update model parameters. Adaptive extraction of key neighbor information is realized through a graph attention mechanism, and adaptability and stability of a formation system in a complex environment are improved by combining dynamic communication range adjustment. The method is suitable for unmanned aerial vehicle group control, robot cooperation and other scenes.

The invention relates to the technical field of computers, and discloses a chip mounter material tracing management method and system based on a knowledge graph. The method comprises the following steps: constructing a dynamic knowledge graph covering the whole process of a patch; injecting multi-source data in real time to drive dynamic updating of the atlas; when the materials are abnormal, reverse tracing and forward influence double-path reasoning are started; isolation, rework and reset instructions are automatically generated and pushed to a manufacturing execution system; and synchronously updating the atlas state to form an auditable record. The system comprises a dynamic knowledge graph construction module, a real-time data injection module, a bidirectional inference engine module, an automatic processing instruction generation module and a graph state synchronization module. According to the scheme, millisecond-level error positioning and closed-loop processing are realized, and the tracing efficiency and the quality guarantee capability are improved.

The invention provides a distributed large-scale anti-traceability elastic network intelligent routing method and system, belongs to the field of network communication and network security, and is suitable for intelligent routing decision optimization in a dynamic network environment. The method is based on a multi-agent reinforcement learning framework, network nodes are mapped into independent agents, and path traceability risks are blocked through local information constraints; dynamic feature aggregation of a neighbor link state is realized by adopting a lightweight graph attention network, and the local sensing efficiency of a large-scale network is improved; a QMIX algorithm is introduced, and network parameters are optimized by nonlinear fusion of a local Q value and global graph state representation through a hybrid network; and in combination with a self-adaptive exploration mechanism driven by action entropy, the sudden change scene strategy response capability is enhanced. According to the system, in military anonymous communication, dark network data transmission and cross-border sensitive services, the anti-traceability and transmission efficiency balance can be remarkably improved, the characteristics of high concealment, high elasticity and low resource consumption are achieved, and a systematic routing solution is provided for a dynamic network environment.

The invention discloses a topological network global optimization method based on graph reinforcement learning, and relates to the field of computational geometry and grid optimizing.The method comprises the steps that a topological optimization process is modeled into a Markov decision process, a topological network containing singular points and angular points is converted into heterogeneous graph representation, and a graph state is constructed; selecting a topology editing action by a reinforcement learning agent fused with the graph neural network based on the current graph state; after executing the action, updating a topological structure and a graph state, and calculating an instant reward based on the global grid quality index change; training intelligent agents in batches through experience playback and time sequence differential learning, so that the intelligent agents learn a mapping strategy from a graph state to an optimal action; and applying the trained policy network deployment to automatic optimization of the new topology. According to the method, full-process automation of topological optimization is achieved, non-local negative effects of local modification are effectively avoided, a global optimal optimization strategy is learned, the grid quality and optimization efficiency are remarkably improved, and the method is suitable for simulation grid generation of aircrafts.

The invention relates to the technical field of computers, in particular to an Internet of Vehicles task unloading optimization method based on a semantic heterogeneous graph and hierarchical reinforcement learning, which comprises the following steps: establishing a vehicle state estimation model, calculating the future position and speed state of each vehicle, and further calculating the remaining connection time between the vehicles; the method comprises the following steps: introducing remaining connection time between vehicles as a dynamic edge weight to construct a multi-semantic heterogeneous graph structure, comprehensively capturing a composite semantic relationship in an Internet of Vehicles task unloading environment, and generating a global state representation vector; establishing a hierarchical reinforcement learning strategy, taking the global state representation vector as an input basis of the hierarchical reinforcement learning strategy, and outputting an optimal unloading path and a target node index; according to the method, through fusion of Kalman filtering link prediction, heterogeneous graph state modeling and a layered Actor-Critic reinforcement learning architecture, the key problems of insufficient link unstable state expression, action space explosion and the like in Internet of Vehicles task unloading are solved.

The invention discloses a multi-aspect reasoning graph method of an NLP interaction model based on reinforcement learning. The method comprises the following steps: extracting entities and relationships from an input text to construct an initial reasoning graph comprising a plurality of reasoning aspects; a reinforcement learning agent is initialized, a state space of the reinforcement learning agent comprises a graph state and user query, and an action space comprises reasoning operation on a graph; the agent updates a graph or generates a path based on a state selection action, and improves performance through continuous interaction based on a comprehensive reward signal optimization policy; according to the method, the limitation that the prior art focuses on a single reasoning dimension is overcome by constructing the multi-dimensional reasoning graph integrating semantics, logic, context and the like and cooperatively operating the sub-graphs by utilizing the reinforcement learning agent, so that the model can carry out all-dimensional and multi-angle deep analysis on complex queries, and the accuracy of the analysis is improved. The reasoning comprehensiveness and the comprehensive judgment accuracy are greatly improved, and therefore more accurate and complete responses are given in tasks such as questions and answers or dialogues.

The invention provides a distributed large-scale anti-traceability elastic network intelligent routing method and system, belongs to the field of network communication and network security, and is suitable for intelligent routing decision optimization in a dynamic network environment. The method is based on a multi-agent reinforcement learning framework, network nodes are mapped into independent agents, and path traceability risks are blocked through local information constraints; dynamic feature aggregation of a neighbor link state is realized by adopting a lightweight graph attention network, and the local sensing efficiency of a large-scale network is improved; a QMIX algorithm is introduced, and network parameters are optimized by nonlinear fusion of a local Q value and global graph state representation through a hybrid network; and in combination with a self-adaptive exploration mechanism driven by action entropy, the sudden change scene strategy response capability is enhanced. According to the system, in military anonymous communication, dark network data transmission and cross-border sensitive services, the anti-traceability and transmission efficiency balance can be remarkably improved, the characteristics of high concealment, high elasticity and low resource consumption are achieved, and a systematic routing solution is provided for a dynamic network environment.

Simulation and validation of neural network systems is provided. In various embodiments, a description of an artificial neural network is read. A directed graph is constructed comprising a plurality of edges and a plurality of nodes, each of the plurality of edges corresponding to a queue and each of the plurality of nodes corresponding to a computing function of the neural network system. A graph state is updated over a plurality of time steps according to the description of the neural network, the graph state being defined by the contents of each of the plurality of queues. Each of a plurality of assertions is tested at each of the plurality of time steps, each of the plurality of assertions being a function of a subset of the graph state. Invalidity of the neural network system is indicated for each violation of one of the plurality of assertions.

The invention provides a double-layer safe trainable quantum machine learning method based on polynomial dynamics Lie algebra, and the method comprises the following steps: 1, constructing core variation simulation meeting the constraint of the polynomial dynamics Lie algebra, and guaranteeing the trainability of a model; 2, performing truncated Chebyshev graph coding at an input end, and constructing a rugged loss function landscape by using graph state entanglement and a Chebyshev tower strategy to prevent a snapshot inversion attack; 3, executing dynamic local scrambling at an output end, applying time-varying random local unitary transformation before measurement, and confusing a linear relation between gradient and a snapshot to prevent recovery attack of the snapshot; and 4, measuring and calculating a loss function, and updating parameters. According to the method, an orthogonal decoupling strategy is adopted, a privacy protection mechanism is externally arranged on an input / output interface, and trainability is anchored to core configuration, so that the capability of resisting algebraic attacks is remarkably improved while model convergence is ensured.

The application provides a forest fire risk intelligent prediction method and system based on a heterogeneous graph state space model, comprising: constructing a unified grid multi-dimensional feature set; the unified grid multi-dimensional feature set comprises feature information of a plurality of grid points distributed in a target area; obtaining a human activity risk index; based on the unified grid multi-dimensional feature set and the human activity risk index, constructing a meteorological-grid heterogeneous graph state space model; based on the meteorological-grid heterogeneous graph state space model, determining the forest fire occurrence probability of each grid point. The method provided in the application constructs a folk custom and agricultural activity dynamic space-time calendar, uses an asymmetric Gaussian space-time kernel function to represent the human fire risk evolution, and reflects the fire risk law of "accumulating before the festival and remaining after the festival"; a physically constrained meteorological-grid heterogeneous graph state space model is constructed, interpolation distortion and error accumulation are avoided, and the forest fire prediction accuracy is improved by coupling double risks through an enhanced step length.

This application discloses a natural language-driven edge-cloud collaborative intelligent operation and maintenance system and its control method, belonging to the field of intelligent operation and maintenance technology. The system includes a management plane, a control plane, an edge agent, an intelligent operation and maintenance center, a task center, and a skill center. The intelligent operation and maintenance center uses a directed graph state machine engine to convert natural language operation and maintenance goals into executable steps. The task center atomically distributes tasks to the edge agent and streams execution fragments. The skill center implements version management and AI-assisted generation of operation and maintenance skills. This invention achieves a fully automated closed loop from understanding operation and maintenance intentions to edge execution and knowledge accumulation, significantly shortening fault repair time, improving operation and maintenance security and knowledge reuse capabilities, and has outstanding substantive features and significant progress.

The invention provides a hardware testing method and device, equipment and a storage medium, and the method comprises the steps: dividing a testing diagram into at least two sub-testing diagrams, each sub-testing diagram comprises a * b regions, and a and b are integers equal to or greater than 2; generating a sub-graph state array, wherein the sub-graph state array is used for recording the test state of the region in the sub-test graph; respectively executing the following processing on one or more sub-test diagrams in the at least two sub-test diagrams to obtain a test result diagram corresponding to the to-be-tested hardware: determining a target area from the sub-test diagrams according to the sub-diagram state array; testing the to-be-tested hardware according to the test parameters corresponding to the target area to obtain a test result, and updating the sub-graph state array; performing filling processing on a target area of the test chart according to a test result; and after the test of each target area in the sub-test chart is completed, initializing the sub-chart state array. According to the method, the memory requirement during testing is reduced by multiplexing the sub-graph state array.

The application discloses a diesel engine whole machine fault prediction method based on a graph neural network, comprising the following steps: collecting multi-source heterogeneous monitoring data of a diesel engine whole machine and key components and dividing the data into initial subgraph data streams; constructing potential fault propagation paths based on the initial subgraph data streams; constructing a nonlinear dependence chain and forming a serialized potential graph; inputting the serialized potential graph into an improved GGSNN model to output a high-dimensional fault representation vector; performing potential weighting deduction based on the high-dimensional fault representation vector and generating a multi-level evolution sequence; performing cross-subgraph path optimization and node priority reordering on the multi-level evolution sequence; and iteratively updating node and subgraph states on the whole machine fault evolution chain to generate a fault prediction result set. The application uses a graph neural network and multi-source data fusion technology to accurately describe a diesel engine fault evolution and risk assessment method, and has the advantages of high prediction accuracy, high real-time performance and quantifiable risk.

The invention discloses a test case generation method and a related product. The generation method comprises the steps that a state machine diagram is obtained, the state machine diagram comprises a plurality of states and a plurality of events used for triggering the states to be converted, the events comprise respective identifiers, and the events with the incidence relation comprise the same identifier; the state machine diagram is analyzed to obtain multiple paths, each path in the multiple paths comprises a starting state, an intermediate state, an ending state and an event used for triggering the starting state, the intermediate state and the ending state to be converted, the multiple states comprise the state included in each path, and the multiple events comprise the event included in each path; and generating a test case based on each path and the identifier of the event included in each path. Therefore, the coverage degree of the test case is greatly expanded, so that the integrity and the accuracy of the test case are improved, and the effectiveness of the test case can be improved.

A quantum computing method includes providing at least one resonator-coupled quantum emitter configured to function as an entangling gate, receiving a plurality of graph states, at least some of the plurality of graph states representing a relationship between qubits therein, and wherein at least one of the qubits in at least two of the plurality of graph states is a photonic qubit, selecting the at least one photonic qubit from each of the at least two of the plurality of graph sates, interacting the selected qubits with the at least one resonator-coupled quantum emitter, and disentangling the at least one resonator-coupled quantum emitter from the selected qubits.

A system and method for measuring, forecasting, and modulating aqueous outflow is described here. The system and method operate on an ocular microphysiological system that reproduces the trabecular-meshwork-membrane-Schlemm's canal interface under a defined hydrodynamic program. The signals may include TEER resistance, pressure-flow measurements, OCT / OCTA images, and Raman spectra. The Ocular MPS system may include encoding the multimodal signals into a device-agnostic feature representation that includes descriptors of junction continuity, belt thickness, tortuosity, pathway activity, and outflow-resistance proxies. The Ocular MPS system may execute a physics-informed graph state-space model that fuses the device-agnostic feature representation into inferred parameters. The inferred parameters may be inferred from barrier integrity, permeability, and outflow facility, subject to monotonic constraints between structural and hydraulic variables. The Ocular MPS system may quantify uncertainty of inferred parameters, generating calibrated forecasts of aqueous outflow performance. The Ocular MPS system may control a drive actuator according to the calibrated forecasts.

It is described a method of performing quantum error correction on a measurement-based quantum computer using fault complexes. A first chain complex is derived based on a first parity check matrix (PCM) and a second PCM that describe a quantum surface code. A second chain complex is derived based on a third PCM of a binary linear code. A fault complex having one or more boundary operators is generated based on a tensor product between the first and the second chain complexes. Qubits on the measurement-based quantum computer are prepared to implement a fault-tolerant graph state based on the one or more boundary operators of the fault complex. Quantum error correction can be performed based on the measurement of the prepared qubits.

This invention discloses a method, system, and electronic device for preparing stable substates based on graph states. The method includes: acquiring and preprocessing stable substate generators to obtain complete stable substate generators; generating an initial parity-check matrix based on the complete stable substate generators; converting the initial parity-check matrix into a graph state parity-check matrix and recording the conversion operations performed during the conversion; constructing a graph state preparation circuit based on the graph state parity-check matrix; constructing a stable substate reduction circuit based on the conversion operations; merging the graph state preparation circuit and the stable substate reduction circuit into a total circuit; acquiring an initial quantum state and executing the total circuit with the initial quantum state to obtain a stable substate. This invention uses graph states as an intermediary, reducing the complexity of circuit design and improving computational efficiency, making it suitable for large-scale quantum systems. Furthermore, the precise mapping relationship between stable substates and graph states avoids approximation errors.

The application provides a hardware test method, device, equipment and storage medium. The method comprises the following steps: dividing a test graph into at least two sub test graphs, wherein each sub test graph comprises a×b regions, a and b are both integers equal to or greater than 2; generating a sub graph state array, wherein the sub graph state array is used for recording the test state of the regions in the sub test graph; performing the following processing on one or more sub test graphs in the at least two sub test graphs respectively to obtain a test result graph corresponding to the hardware to be tested: determining a target region from the sub test graph according to the sub graph state array; testing the hardware to be tested according to the test parameters corresponding to the target region, obtaining a test result, and updating the sub graph state array; performing filling processing on the target region of the test graph according to the test result; and initializing the sub graph state array after completing the test on each target region in the sub test graph. The method reduces the memory requirement during the test by reusing the sub graph state array.

A system and method for self-executing graphs wherein execution semantics are encoded within graph elements themselves rather than imposed by external schedulers. The system comprises a dynamic property graph with vertices and edges that encode execution semantics specifying computational operations and graph-internal triggering conditions. An execution engine evaluates these triggering conditions by monitoring graph state, detects satisfaction through graph-internal evaluation, and initiates bound computational operations. Triggering conditions are expressed in terms of vertex or edge traversal, property changes, geometric properties such as curvature, or topological connectivity patterns. The graph autonomously determines when and which operations execute based on graph-resident execution semantics. In distributed embodiments, multiple local coordinators evaluate triggers within assigned graph regions and coordinate through peer-to-peer messaging without centralized scheduling. Applications include autonomous cognitive systems, adaptive control systems, and scalable distributed computation requiring self-regulating execution without centralized bottlenecks.

The application discloses a kind of virtual space multi-modal dynamic construction and collaborative intervention method for digital healing, its characteristics are that the method specifically includes: constructing multi-modal emotional knowledge graph containing collaborative constraint;Based on the multi-modal benchmark data of user neutral emotional state, the knowledge graph is personalized calibrated;According to user real-time multi-modal data, activate graph state node, use graph path reasoning algorithm and dynamic collaborative gating mechanism to calculate optimal collaborative intervention path and generate dynamic digital prescription;It is parsed into multi-modal rendering instructions to drive virtual engine synchronous rendering, and according to biological feedback, real-time update graph edge weight and subsequent collaborative intervention path.Compared with prior art, the application can realize the collaborative fusion and dynamic intervention of multi-modal information, improve the consistency, adaptability and stability of virtual space feedback regulation, reduce the perception conflict caused by heterogeneous modal mismatch, achieve the effect of digital healing, and has good application prospect.

The invention provides an intelligent calculation operation and maintenance RDF knowledge graph updating method and device and electronic device.The method comprises the steps that in response to a detected dynamic change event associated with a target object, the dynamic change event is converted into a resource description framework RDF triple and an operation intention corresponding to the RDF triple; according to the event type of the dynamic change event, performing target verification on the initial RDF knowledge graph of the intelligent calculation operation and maintenance; and under the condition that the initial RDF mapping knowledge domain passes the target verification, updating the initial RDF mapping knowledge domain according to the RDF triple and the operation intention to obtain an updated RDF mapping knowledge domain. According to the method, the consistency and integrity of graph state evolution can be ensured, data exception caused by conflict change can be effectively avoided, and the reliability of knowledge graph maintenance and the intelligent calculation operation and maintenance level are improved.

A method and a device for correcting quantum state information are disclosed. A decoder receives information identifying syndrome values for a plurality of entangled qubit states represented by a graph state with a respective edge of the graph state corresponding to a respective qubit state of the plurality of entangled qubit states. The decoder repeats identifying one or more clusters of qubit states and / or syndrome states in the graph state until all of the one or more identified clusters are determined to be valid while increasing a size of a respective cluster each time the identifying operation is performed. The decoder reconstructs one or more qubit states and / or syndrome states for respective clusters; and stores information identifying the one or more reconstructed qubit states and / or syndrome states.

The embodiment of the invention provides a task scheduling method and device for a graph processing request. The method comprises the steps of obtaining a plurality of task units corresponding to a target image processing request; the plurality of task units correspond to a plurality of calculation modes; obtaining a state machine migration graph which is uniformly constructed for multiple calculation modes; the state machine migration graph comprises a plurality of nodes, each node indicates an execution state of a subtask split from each calculation mode, and a migration edge between the nodes corresponds to the subtask; according to the state machine migration graph, the task units are scheduled and executed, and the scheduling execution of any target task unit comprises the steps that in a first state indicated by the first node, in response to a first triggering condition for starting to migrate to a second state, a target job in the target task unit is scheduled and executed, therefore, the current state is transferred to the second state; the target job belongs to a target subtask corresponding to a migration edge between the first node and a second node corresponding to the second state.

The application relates to a multi-unmanned aerial vehicle assisted task graph calculation unloading method and device with a graph structure enhancement. The method comprises the following steps: constructing a multi-unmanned aerial vehicle assisted mobile edge calculation system; modeling a calculation unloading problem as a cooperative multi-agent semi-Markov decision process; constructing a graph structure enhanced MAPPO collaborative strategy learning model, including local graph observation construction, a graph structure enhanced Actor network and a centralized graph Critic network; generating a local graph observation at each decision time; the Actor network outputs a hybrid action of subtask selection, execution position and power control; after the action is executed, a total system cost function is calculated, and a regret type shared reward is obtained based on a cost difference of adjacent decision steps; the Critic network receives a shared graph state to evaluate a global value; and the network is iteratively updated by combining the shared reward with a value function loss of a PPO pruning target. The method can effectively reduce the total system time delay overhead and stabilize convergence.

The invention discloses a pose-driven controllable video generation method based on a diffusion model, and the method comprises the following steps: obtaining a reference appearance image set and a pose sequence, generating an observation tensor, and building a skeleton template; calling a micro SoftPOSIT to obtain a soft distribution matrix, a frame-level pose initial result, a visibility weight and a matching confidence coefficient; alternately updating and applying bone length invariance, joint amplitude limiting and kinematic chain consistency constraint, and outputting frame-level pose estimation; organizing an initial attitude sequence according to the matching confidence, and generating a visibility weight sequence and a soft distribution matrix sequence; establishing a cross-frame factor graph, outputting a time consistent attitude sequence and generating a factor graph state; integrating the quantities into a video condition sequence; in diffusion denoising, temperature parameters and noise intensity are synchronously adjusted and matched, attitude consistent guidance is calculated, and the state of a factor graph is incrementally updated; and generating a video frame sequence according to posture consistency guidance. According to the method, the attitude consistency and the time sequence stability are improved, and the controllability is enhanced.

A single-shot graph state generator prepares matter qubits in an entangled graph state. Each matter qubit may be in a first state, a second state, or an excited state. The single-shot graph state generator drives the matter qubits with one or more pulses that cause a transition from the first state to the excited state. A decay of a matter qubit from the excited state to the first state generates a corresponding photonic qubit. The single-shot graph state generator measures the matter qubits to teleport the entangled graph state to the photonic qubits.

To solve problems in sales demand prediction for smart phone and suitable inventory calculation, and automatic output of a charted analysis result by input of text.SOLUTION: A system includes: means to automatically calculate sales demand prediction for smart phones and suitable stock along with it by having ChatGPT learn data of data warehouse (DWH); and means to output an analysis result in a graph state by inputting information needing analysis by text.SELECTED DRAWING: Figure 1

A single-shot graph state generator prepares matter qubits in an entangled graph state. Each matter qubit may be in a first state, a second state, or an excited state. The single-shot graph state generator drives the matter qubits with one or more pulses that cause a transition from the first state to the excited state. A decay of a matter qubit from the excited state to the first state generates a corresponding photonic qubit. The single-shot graph state generator measures the matter qubits to teleport the entangled graph state to the photonic qubits.