12results about How to "Optimize execution efficiency" patented technology

The invention provides a rapid construction and reuse method and system for an application framework, and relates to the technical field of application framework construction, and the method comprises the following steps: mapping code snippets to a high-dimensional tensor space, extracting multi-level code features, and constructing a code association graph and a deep semantic network; constructing a hierarchical Bayesian network based on the spatial-temporal feature map, executing variational reasoning and adaptive particle filtering, and generating a reconstruction code; driving a multi-level generation network by using a dual-dimension discriminator, optimizing a code structure and a dependency relationship, and generating a final optimization code through program slicing and incremental synthesis; according to the method, rapid construction and reuse of the application framework are realized, and the code quality and the execution efficiency are improved.

The invention relates to the field of software engineering and computer science, in particular to an intelligent code optimization and reconstruction system based on a large model, which adopts a comprehensive code analysis module to perform static and dynamic analysis and identify potential problems and complexity indexes in codes. Through an optimization suggestion generation module, personalized optimization suggestions are generated by using a neural network algorithm, and automatic code reconstruction is realized for different programming languages. And the system ensures the functional integrity of the reconstructed code through a reconstruction verification module, and provides a user feedback mechanism to optimize a subsequent strategy. Finally, the readability, maintainability and execution efficiency of the codes are remarkably improved, and the intelligent code optimization effect is achieved.

The invention belongs to the field of nursing robots, and particularly discloses a multi-robot collaborative operation method and system for home nursing of old people, and the method comprises the steps: building a virtual simulation platform which comprises a simulation robot, the old people and a home environment, a large motion planning model built based on the PaLM-E is deployed on the simulation robot; the multi-modal information comprises visual sense, touch sense, sound and temperature sensing information; multi-robot cooperative operation training is carried out based on a virtual simulation platform, so that the action planning large model is trained; multi-modal information and nursing task information about the old people are obtained and input into the trained motion planning large model, unit motion planning of all robots is obtained, and home nursing of the old people with multi-robot collaborative operation is achieved. According to the invention, personalized, efficient, comprehensive, safe and intelligent nursing services can be provided for old people.

The invention discloses a method, system and related product for performing release management on an operation activity, and the method comprises the steps: determining a plurality of agents for performing release management on the operation activity and release management role information corresponding to each agent according to a standard operation process of the operation activity, a putting management process of a plurality of intelligent agents is set, and the intelligent agents are connected with a first robot used for executing operation; performing role debugging on the plurality of agents according to the release management role information and the release management process to obtain the plurality of agents after role debugging; and realizing delivery management of the operation activity according to a delivery management process by using the plurality of intelligent agents subjected to role debugging and the robots connected with the intelligent agents. By utilizing the scheme disclosed by the invention, the personalized group operation service can be realized, and various operation operations can be executed in the user group, so that the group operation effect is improved.

This invention, belonging to the field of computational processing technology, proposes a method to accelerate sparse 3D convolution based on thread bundle alignment and memory access rearrangement. To address the issues of discontinuous memory access and low memory bandwidth utilization caused by the sparsity of convolution kernels in GPU architectures, this invention establishes a thread bundle aligned convolution kernel format method based on the direct convolution method. This method uses the width of the thread bundle as the basic unit for grouping and compression to obtain the thread bundle aligned convolution kernel format. Considering the single-instruction multithreading (SIMT) characteristics of graphics processors and the properties of convolution operations, this method is used to merge sparse convolution kernel data and perform vectorized memory access. A memory access conflict resolution rearrangement method is also established: Input data is stored in shared memory using a DHWC layout, and then a greedy algorithm is used to rearrange the weight storage order within each basic unit of the WAF, thereby eliminating shared memory conflicts for input data access by manipulating the address of each memory access.

The application is suitable for the field of artificial intelligence technology, and provides an agent skill dynamic arrangement method and system based on task semantic decomposition, comprising the following steps: receiving a task request, performing intention structured extraction, and generating a structured task description; according to the structured task description, calculating the matching degree score of each skill and the task intention from a skill library, and outputting a candidate skill set; performing double-layer semantic matching verification on the candidate skill set, and outputting a skill list; extracting semantic constraints in the structured task description, including spatial constraints, time constraints and business logic constraints, and generating constraint structured representation; based on the skill list and the constraint structured representation, analyzing the dependency relationship and parallel execution possibility among the skills, optimizing the execution order according to geographical context information, and constructing a directed acyclic graph of task execution. The application eliminates the problem of incompatible skill combination or missing function from the semantic level, and can minimize the task execution path length.

The invention discloses a rapid topology analysis method for single-ended element switch deflection, which comprises the following steps: acquiring a CIM file, analyzing the CIM file into a vertex and edge data set oriented to a graph database, and constructing a node switch graph model of a power grid; solving a network topology by applying a graph division technology to construct a bus branch graph model; acquiring real-time equipment state data, updating the node switch graph model to a real-time state, and traversing associated equipment; if the associated equipment is a single-ended element, generating a first topological node, and mapping the first topological node to a bus branch graph model node by using an optimized graph topology analysis algorithm; if other functional elements are associated, a traditional topology analysis method is used for judgment, a second topology node is generated, and the second topology node is mapped to a bus branch diagram model node; and mapping the first topological node and the second topological node in the bus branch diagram model to the corresponding transmission line and the series compensator node, thereby providing accurate topological information for the operation state of the power grid.

The present application relates to a kind of rail transit line congestion train operation control system and method, comprising: step 1, according to the delay time of congestion, determine the each train in the range of affected area;Step 2, introduce travel speed-interval calculation model, obtain the slow-down multi-stop control time of each train for each train in the range of affected area;Step 3, slow-down multi-stop control time is automatically issued to each train to execute.The system includes intelligent scheduling strategy operation module and intelligent scheduling strategy decomposition execution module, for running travel speed-interval calculation model, dynamically calculate the slow-down multi-stop control time of each train, and output to each train for execution.The system relies on algorithm to automatically control the automatic adjustment of all trains in upstream and downstream direction, without scheduling manual control, and execution efficiency is excellent.

The invention belongs to the technical field of terminal testing, and particularly relates to a mobile intelligent terminal compatibility testing method which comprises the following steps: S1, aggregating terminal hardware configuration, software environment and network state data through federal learning, and constructing a real-time testing matrix; s2, synthesizing an edge scene test case based on the generative adversarial network; s3, synchronously collecting a touch track, a display frame rate and an audio synchronization deviation, and quantifying user interaction experience in combination with eye tracker data; s4, deploying a lightweight virtualization intermediate layer between the gap / Android / iOS, automatically converting an equipment interconnection instruction, and verifying cross-chain compatibility; and S5, tracking the privacy data flow by adopting dynamic taint analysis, and detecting illegal behaviors in real time. According to the method, through dynamic environment modeling, real-time test matrix construction and hierarchical aggregation architecture, terminal nodes encrypt and upload local data by using differential privacy and homomorphic encryption technologies, so that the security of the data is ensured.

The invention discloses a machine learning method and device, computer equipment and a storage medium, and belongs to the technical field of artificial intelligence. The environment parameters are collected according to the preset time interval, the corresponding operation instruction is generated based on each parameter, and the intelligent agent is instructed to obtain the operation result after execution; comparing with a preset standard result, dividing the operation result into a correct first operation result and a wrong second operation result, respectively extracting environment parameters and operation instructions matched with the correct first operation result and the wrong second operation result, and constructing a training data set containing the first environment parameters, the first operation instructions, the second environment parameters and the second operation instructions. And performing iterative optimization training on the intelligent agent by using the training data set, and finally obtaining a target intelligent agent capable of executing the target operation instruction and generating an efficient result. The invention further relates to the technical field of block chains, and the environment parameters are stored on a block chain network. According to the invention, through environment parameter dynamic monitoring and a result feedback mechanism, the learning ability and adaptability of the intelligent agent are significantly improved.

The application discloses a multi-model reasoning method and device based on a graph structure, a medium and a program product, wherein the method comprises the following steps: querying a task target node in a graph database and starting from the task target node, and performing reasoning path expansion in the graph database according to a corresponding driving model, wherein the graph database comprises a plurality of directed acyclic graphs, and each directed acyclic graph is obtained by mapping metadata of a registered analysis model; generating a task topology execution graph according to model nodes involved in the expanded reasoning path and the corresponding driving model; and sequentially scheduling actual execution of each model node according to the task topology execution graph, and obtaining an explanation result corresponding to the task target node. The application improves the reasoning execution efficiency by explicitly attributing the analysis model and optimizing the reasoning path by using graph calculation, and improves the intelligence, performance and interpretability of multi-model collaborative analysis.

The application provides a kind of multi-model API docking management method and system based on deep learning, it is related to artificial intelligence and cloud computing technology field, including the following steps: collect and standardize all the AI prompt word templates and skill assets generated in workflow storage, to generate AI prompt word template library and skill asset library;Request data is optimized;The state of the multi-business model is evaluated in combination with the optimized request data to determine the target business model, and the optimized request data is distributed to the target business model to perform inference operations;Real-time acquisition of inference results and performance indicators generated during the execution of the target business model, and the inference results and performance indicators are returned to the cloud storage space;The AI prompt word template library and skill asset library are updated to realize the docking optimization of multi-business model.The application realizes the efficient, low-cost and adaptive optimization of multi-business model call, with good scalability and practical value.