5765results about "Program loading/initiating" patented technology

The invention provides a large model lightweight reasoning deployment method under limited hardware resources, and the method comprises the steps: quantifying the weight importance of a large model through a composite index of gradient sensitivity and activation frequency, and carrying out pruning operation in combination with an improved index weighted moving average strategy, thereby obtaining a structured sparse model; the sparse model is divided into sub-networks by adopting double rules, a routing decision network is trained, and an adaptive feature shunting architecture model is constructed; a multi-precision weight set is generated through a nested quantization technology, quantization bit width is dynamically adjusted, and edge equipment hardware parameters are adapted to complete reasoning environment initialization; after a reasoning request is received, an optimal sub-network is selected based on the trained routing decision network, corresponding weights are loaded in parallel, and a reasoning result is fused and output; and converting a reasoning result format, and dynamically optimizing a scheduling strategy based on a system real-time monitoring index. The method is compatible with a mainstream large model and a hardware platform, and an efficient and universal deployment scheme is provided for end-side AI engineering landing.

Systems and methods are described for building artificial intelligence (“AI”) agents. A server can provide a user interface (UI) that includes options to select and connect various agent objects. The agent objects can include prompt objects, dataset objects, model objects, and one or more code objects. A subset of the agent objects can be identified as available for selection based on evaluation of at least one management policy associated with an administrative user. A manifest file is generated based on the connected agent objects, and the manifest is validated against dependency rules to ensure that the stages of the agent meet prerequisites for the stages. Then, the server performs a simulated execution of an agent that corresponds to the validated manifest file, including an identification of at least one execution metric associated with the simulated execution.

The invention provides an online map tile fast loading and caching optimization method, medium and system, and belongs to the technical field of map processing.The method includes the steps that a tile index database and an exchange priority matrix are established, a viewport moving trend vector is calculated according to user sliding operation, a future viewport range is predicted, and a preloaded tile list is screened; the method comprises the following steps: performing cache validity check on tiles, establishing a small change detection vector to judge whether updating is needed, establishing a tile hit rate statistical matrix, fusing a time decay factor and a spatial correlation factor to calculate a popularity score, and gathering high-popularity tiles in a quick access area by adopting a thermodynamics-imitating entropy reduction algorithm to realize cache ordering. And the next tile demand of the user is predicted, so that the technical problem that the overall loading response speed is reduced due to an overlong high-value tile access path caused by disordered distribution of cache resources in the online map tile loading process is solved.

Systems and methods are described for dynamic execution of artificial intelligence (“AI”) agents. A server can receive, from a client device, an input associated with an AI agent. Based on a manifest file or user profile, the server can identify a management policy that applies to the AI agent. The server then dynamically configures access to the agent objects based on applying the management policy. The management policy is applied to a device status of the client device, a user profile of a user of the client device, and / or a network configuration of the client device. The server then executes a modified workflow based on the dynamically configured access, wherein the modified workflow bypasses or changes operation of at least one of the agent objects. Based on the modified workflow, the server transmits an output to the client device.

Systems and methods are described for building artificial intelligence (“AI”) agents. A server can provide a user interface (UI) that includes options to select and connect various agent objects. The agent objects can include prompt objects, dataset objects, model objects, and one or more code objects. A subset of the agent objects can be identified as available for selection based on evaluation of at least one management policy associated with an administrative user. A manifest file is generated based on the connected agent objects, and the manifest is validated against dependency rules to ensure that the stages of the agent meet prerequisites for the stages. Then, the server performs a simulated execution of an agent that corresponds to the validated manifest file, including an identification of at least one execution metric associated with the simulated execution.

The invention relates to the technical field of data processing, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a data processing and analysis method, device, equipment and medium, and the method comprises the steps: obtaining a data source set, and registering a source label and field mapping relation; executing extraction, cleaning, conversion and loading in the data processing assembly line based on the data source set, the source label and the field mapping relation to generate a processing table set; performing quality verification and exception handling on the processing table set to generate a verified processing table set; generating an index cloud based on the verified processing table set and the configuration strategy; the index cloud is pushed to the analysis tool set through the interface transmission channel, and the analysis tool set generates analysis output according to the index cloud. According to the method, through unified data registration, standardized processing and quality verification, the index cloud is generated in combination with the configuration strategy and pushed to the analysis tool set, the analysis result is generated and transmitted, the data consistency is ensured, and the real-time performance is improved.

The invention discloses an automatic compiling and adapting method for an RISC-V extension instruction, and belongs to the technical field of compilers. According to the method, a dynamic DSL (Digital Subscriber Line) for registering a custom instruction, adding register use constraints and defining a specific code mode is designed. The LLVM plug-in is used for automatically integrating the registered custom instruction in the compiling process. The invention relates to a self-defined instruction rapid adaptation mechanism which is specially provided for a continuously provided RISC-V self-defined instruction expansion scene, and by automatically generating correct assembly codes of registered self-defined instructions, correct registers are automatically distributed and reserved according to register use constraints; the method comprises the following steps of: searching a defined specific code mode and semantics, and automatically generating and inserting a registered custom instruction in a program, so as to realize the rapid adaptation of a compiler to an RISC-V custom extension instruction and the automatic use of a program code to the custom instruction, and the whole process does not need a developer to manually modify the compiler or a program source code.

The invention relates to the technical field of audio control, and discloses a low-delay audio input switching method and system, a storage medium and equipment, and the method comprises the steps: carrying out the parallel pre-initialization of a plurality of pieces of audio input equipment, and building an independent parallel audio data cache for each piece of equipment; monitoring the state of each audio input device and the audio stream quality in real time, and judging whether switching is triggered or not based on a multi-factor decision model; after the switching decision is triggered, seamless audio data stream switching is executed, and format unification processing and cross fade-in and fade-out transition are included; a unified equipment operation interface is provided through the hardware abstraction layer, and system resources are optimized and managed; a delay sensing closed-loop control mechanism is constructed, processing delay of each link is monitored in real time, a caching strategy, a processing algorithm and resource allocation parameters are dynamically adjusted, self-adaptive balance of low delay and high tone quality is achieved, and through the method, quick and smooth switching of audio input equipment is achieved, delay is remarkably reduced, and real-time audio experience is improved.

Systems and methods are described for dynamic execution of artificial intelligence (“AI”) agents. A server can receive, from a client device, an input associated with an AI agent. Based on a manifest file or user profile, the server can identify a management policy that applies to the AI agent. The server then dynamically configures access to the agent objects based on applying the management policy. The management policy is applied to a device status of the client device, a user profile of a user of the client device, and / or a network configuration of the client device. The server then executes a modified workflow based on the dynamically configured access, wherein the modified workflow bypasses or changes operation of at least one of the agent objects. Based on the modified workflow, the server transmits an output to the client device.

The invention relates to the technical field of computers, and discloses a deployment strategy adjustment method and system, electronic equipment and a storage medium, and the method comprises the steps: configuring a multi-dimensional index set, collecting and processing dynamic data, generating a deployment adjustment strategy based on the synergistic effect of a dynamic weight distribution algorithm and a prediction model, and executing the deployment adjustment strategy. And selecting a progressive adjustment mode or an aggressive adjustment mode based on the change amplitude of the index in the processed dynamic data to carry out resource or configuration change, monitoring a corresponding state after execution of the deployment adjustment strategy to generate feedback data, and when the feedback data exceeds a preset recovery threshold, recovering to a state before execution of the deployment adjustment strategy. And the adjustment effect is quantitatively evaluated based on feedback data, and a dynamic weight distribution algorithm and prediction model parameters are optimized. According to the method, the problems that a traditional deployment strategy depends on manpower and response lags behind are solved, and the operation and maintenance efficiency and the service toughness can be remarkably improved in a network request high-concurrency scene.

The invention relates to the technical field of industrial automation process control, and discloses a data source adaptation and control binding method, device, equipment and medium, comprising: establishing a variable model to generate a variable directory, constructing a data source adapter to generate a data source mapping table and a read-write interface set, generating a control tree in a visual control designer, establishing a binding relationship between control attributes and variables, loading a variable binding engine, analyzing a binding path, generating an execution plan, registering change monitoring, accessing a read-write interface set of a target data source adapter based on the execution plan to complete variable read-write, and generating an execution record, and responding to a variable change event or a control input event to trigger control value write-back and interface refreshing, and finally collecting a variable directory, a binding relationship, a control tree and an execution record to generate a package configuration file. Through a unified variable model and control binding mechanism, cross-data-source analysis and bidirectional linkage are realized, and the interaction efficiency and the system maintainability are improved.

The invention relates to the technical field of data processing and data quality management, and provides a data quality verification and cleaning method and system based on dynamic rule configuration.The method comprises the steps that a rule configuration interface is provided, and verification rules are stored in a rule base in a configurable metadata form; receiving initial data from the multi-source data interface and performing format conversion to obtain structured data; dynamically loading a target verification rule from a rule base according to the attribute of the structured data, and executing verification to identify and mark abnormal data; according to the abnormal type of the abnormal data and the association rule ID, calling a cleaning strategy bound with the service scene to carry out differentiation processing; and outputting the cleaned effective data and the exception handling report. According to the mode, the adaptability and processing efficiency of data processing are enhanced while the data quality is improved.

Provided are a multi-agent construction method and apparatus, a task processing method and apparatus, a computing device, a storage medium, a computer program product, and a chip. The multi-agent construction method comprises: splitting a task into a plurality of sub-tasks; determining, from an agent pool, agents respectively matching the plurality of sub-tasks, wherein the agent pool comprises a plurality of agents, and the agents matching the sub-tasks can process the sub-tasks by calling at least one of tools and other agents which are configured by the agents; and constructing a multi-agent system on the basis of the agents respectively matching the plurality of sub-tasks, wherein the multi-agent system is used for processing the task.

The invention relates to the technical field of computers, in particular to a dynamic configuration initialization and version synchronization method and system for multi-agent service, and the method comprises the following steps: obtaining a database update lock, constructing a database migration script dependency relationship, determining a current version of a database mode, upgrading and deploying the database mode, and verifying agent versions. The method has the beneficial effects that a rigorous database updating lock control mechanism, perfect migration script dependency chain verification, agent version management, port availability detection, idempotence and transaction safeguard measures are introduced, so that the processes of database structure upgrading and agent version initialization in a distributed deployment environment can be effectively supported; and the data consistency, the operation stability and the service continuity of the system are ensured.

Methods, systems, products, services, and apparatuses for generative AI systems configured to produce computer program code using a knowledge base that includes a client-specific code base, including: receiving one or more input tokens associated with a computer program; accessing information describing a domain-specific codebase; and generating, based on the one or more input tokens associated with the computer program and information describing a domain-specific codebase, suggested code to insert into the computer program.

Systems and methods are described for tool discovery and ingestion for artificial intelligence (“AI”) agents. An AI platform can discover a first tool specification that includes an action and a description. The first tool specification is ingested to create a first tool object. The first tool object includes the action and an endpoint. Tool labels are determined from the specification and applied to the first tool object. A user interface displays the tool object, and it is added to an AI agent. The AI agent includes a manifest file that is used to execute the AI agent. This includes determining whether the AI agent is authorized to perform the action, and providing the AI agent with access to a tool credential, wherein the tool credential is sent to the endpoint.

The invention discloses an implementation system, method and device for a multi-source heterogeneous data connector in a trusted data space and a storage medium, and relates to the technical field of data processing, the implementation system for the multi-source heterogeneous data connector in the trusted data space comprises a data source access layer, a data analysis layer, a data conversion and mapping layer and a data output layer; wherein the data source access layer is used for receiving a data source access request of a user and establishing connection with a target data source according to the data source access request; the data analysis layer is used for acquiring original data from the target data source and analyzing the original data according to a preset analysis strategy to obtain preliminarily analyzed data; the data conversion and mapping layer is used for processing the preliminary analysis data according to a semantic mapping rule to generate standardized data; and the data output layer is used for outputting the standardized data to a specified target application system through a standard interface.

The embodiment of the invention provides an AI large model training resource dynamic scheduling method and device based on a VGPU, and effective planning of resources is realized through innovatively designing a resource prediction system and through index analysis and demand prediction. A multi-target scheduling mechanism is constructed, and a reliable scheduling strategy is established in combination with competition analysis and scheme screening. Virtualized isolation is introduced, and the training stability is ensured through VGPU and resource control. According to the method, the defects of the traditional technology in the aspects of resource prediction, scheduling optimization, virtualization isolation and the like are effectively overcome, and technical support is provided for large model training.

A method for automated control code generation and verification includes: receiving a natural language command, the natural language command being configured to instruct the large language model to output a code text that meets control requirements corresponding to the natural language command; performing matching retrieval on a vector database according to the natural language command to obtain a sample code snippet; obtaining API structured information corresponding to the sample code snippet from a knowledge graph database; generating an initial control code according to the sample code snippet and the API structured information; and performing a multi-level virtual operation verification on the initial control code in a software motion control system, and generating a target control code according to multi-level verification results confirmed multiple times by the user and the initial control code.

The invention provides a cloud edge-end collaborative multi-modal data distributed governance method, which comprises the following steps that: an edge node obtains multi-modal time sequence data, adopts Gaussian process regression to carry out probability interpolation, and obtains a multi-modal time sequence based on a heterogeneous space association graph which is pre-constructed at a cloud end and contains geomechanical parameters; the method comprises the following steps: extracting a multi-modal space fusion feature sequence through a space geological attention map neural network, inputting the multi-modal space fusion feature sequence into a hierarchical cross attention Transform network, and generating time sequence features at edge nodes by an intra-modal coding layer; the cloud decides a calculation position of an inter-modal fusion layer according to a real-time load, network time delay and an early warning level: if edge execution is carried out, a full-modal fusion feature is generated and uploaded; if the cloud executes the operation, the data is transmitted to the cloud for fusion, the cloud decodes the full-modal representation prediction multi-measuring-point state, the uncertainty function of the graph structure is optimized through the residual covariance, and the association graph weight and the geomechanical parameters are iteratively updated.

The invention discloses an eBPF-based kernel optimization method, system, equipment and medium, belongs to the technical field of Linux kernel network optimization, and aims to solve the technical problem of how to improve the transmission performance for a specific data stream and specify the data stream to carry out a special routing and transmission process so as to provide services for a special network adapter and improve the transmission performance of the specific data stream. According to the technical scheme, the method comprises the steps that data packet interception is carried out on a plurality of kernel eBPF loading points and program interception points selected by a user through an eBPF kernel hook program, routing query is carried out with a routing kernel module, and selection of an operation function and routing parameters is completed; the routing kernel module completes a routing decision based on various complex routing requirements, configuration conditions, data packet conditions and environment conditions, and returns a query result to the eBPF kernel hook program, and the query result is executed by the eBPF kernel hook program. The system comprises an eBPF user mode management program module, a user mode control tool, an eBPF kernel hook program module and a routing kernel module.

The invention discloses an Internet of Things edge computing task unloading method based on a comprehensive gateway, which belongs to the technical field of computer data processing, and comprises the following steps: acquiring tasks, equipment and network data of Internet of Things equipment, and fusing to generate a unified information view; determining task priorities based on the view by applying a dynamic priority evaluation mechanism, and generating a dynamic priority list; applying the time sequence analysis model to generate short-term prediction data; in combination with the dynamic priority list and the short-term prediction data, multi-objective optimization is performed through a collaborative decision model, and a task unloading strategy is generated; generating feedback data according to the task unloading strategy; and generating an adjustment signal by using the feedback data, and updating the dynamic priority evaluation mechanism and the collaborative decision model. According to the method, the unified information view is generated by fusing the multi-source data, collaborative decision making is carried out in combination with dynamic priority evaluation and short-term prediction data, and the model is continuously optimized by utilizing feedback data, so that self-adaptive, forward-looking and efficient task unloading can be realized.

The invention discloses a middleware management method and device, relates to the technical field of middleware, and mainly aims to realize unified management and configuration of middleware, reduce the management and maintenance difficulty of the middleware and improve the flexibility and convenience of flow management. According to the main technical scheme, each middleware instance is divided into one or more logic clusters according to operation characteristics of each service, wherein the operation characteristics are used for representing characteristics related to high-reliability operation of the services; general configuration setting is carried out on the one or more logic clusters to obtain global configuration information containing all the logic clusters, and the general configuration setting at least comprises setting of physical mapping and routing strategies of middleware instances docked by all the services in the running process; and respectively issuing the global configuration information to each middleware instance contained in each logic cluster so as to run according to the global configuration information when a service butted with the middleware instance contained in each logic cluster is started. The middleware management method and device are used for middleware management.

The invention relates to a visual component management method and system based on support of multi-user collaboration, and belongs to the technical field of computer software collaborative design. Wherein multi-party editing requests are received and verified through a cooperative access mechanism, structured component data are constructed by using a component modeling model, concurrent control and a difference buffer are adopted to coordinate parallel modification, write conflicts are identified and processed based on a conflict detection algorithm, and traceable management of component versions is realized through a snapshot generation model. According to the method, efficient, consistent and conflict-controllable unified management of the visual components in a multi-person online cooperation scene is realized, and the cooperation efficiency of component development and the system reliability are effectively improved.

The system provides a permissions engine derives permissions for responses from artificial intelligence (AI) agents. The permissions are derived based on the access permissions of chunks used to generate the response, and the response is cached for reuse with users meeting the permissions level. The system can also derive an access permission level for a user attempting to access the cached result or other content. This can be based on a managed access profile that includes user behavior criteria and device criteria. Likewise, the system can validate and derive sensitivity levels of resources that are ingested for use in the artificial intelligence (AI) agents and discriminate between chunks to use in synthesizing results based on the derived permissions of the chunks and their relationship to the user.

The invention provides a processing method, system and device for expanding an OpenResty request based on a WASM plug-in and a medium, and belongs to the fields of network communication, application gateways and the like, and the method specifically comprises the steps that in the starting stage of an OpenResty working process, a WASM running environment supporting a WASI interface is initialized and embedded; the method comprises the following steps: dividing execution stages of an OpenResty request processing flow, and setting a hook point in each stage; a WebAssessment plug-in module is called, and the WebAssessment plug Executing access control, security detection or request filtering processing according to the input data, and generating a processing result containing an allowing, intercepting or modifying instruction; and the plug-in management module judges whether to continue to execute a subsequent OpenResty processing flow or not according to the received processing result, and supports hot updating or unloading operation on the WebAssembly plug-in module when the configuration is changed. By introducing a WASM plug-in mechanism, the OpenResty is endowed with higher expandability and flexibility in the aspects of flow processing, access control, security detection and the like.

The invention relates to the technical field of data security, in particular to a data desensitization method and system based on combination of an MCP protocol and rule driving. The method comprises the following steps: loading a service rule through an MCP Server and registering a desensitization tool as a standardized tool; user query is received through the MCP Client, and user intention is analyzed by the AI agent; the AI agent dynamically calls a standardized tool to form a desensitization task chain according to a user intention and a business rule; and executing the desensitization task chain to process the target data, wherein the calling process comprises dynamic strategy adjustment based on context information. The system comprises an MCP service layer, a desensitization tool library module and an intelligent decision and integration layer. Through MCP protocol standardization integration, service rule driving and AI agent decision making, dynamic weight distribution and probabilistic algorithm selection of a desensitization strategy are achieved, the flexibility, the intelligent level and the system integration efficiency of data desensitization are effectively improved, and meanwhile the safety and compliance guarantee capacity is enhanced.

A computer implemented method is provided. The computer implemented method includes receiving an intermediate representation of a source code, intentionally injecting a weak code path at a point within the intermediate representation to create a modified intermediate representation, performing a path profiling on the modified intermediate representation to generate a particular path identifier for each path within the modified intermediate representation, and identifying the particular path identifier of the weak code path for use by a monitoring system. A monitoring system is also provided. The monitoring system monitors an executable code during runtime for execution of a path having a particular path identifier corresponding to the injected intentionally weak code path.

The invention belongs to the field of mobile edge computing, and particularly relates to a resource pricing problem based on a Stackelberg game in mobile edge computing. The method comprises the following steps: constructing a mobile edge computing model, wherein the mobile edge computing model comprises a server base station, task equipment and cooperative equipment; in a task execution period, for a server base station profit maximization problem of resource pricing based on a Stackelberg game in mobile edge computing, delay brought by task unloading is fully considered in problem modeling, and the delay requirement of a task is met to the maximum extent; analyzing and establishing a dynamic punishment and redistribution mechanism for task unloading failure of the cooperative equipment, wherein the mechanism comprises a punishment cost calculation module, a task value reevaluation module and a game strategy updating module; and obtaining an unloading strategy of the task equipment, a task allocation strategy of the server base station and a resource pricing strategy after the server base station and the cooperative equipment carry out the Stackelberg game by utilizing a greedy strategy, thereby obtaining a final task allocation scheme for maximizing the profit of the server base station. According to the method, the optimal task execution and resource allocation strategy of the server base station is formulated, and the relationship among the task value, the resource cost and the additional resource purchase cost is balanced as required, so that more flexible and efficient task scheduling service can be provided.

The invention discloses a risk control strategy-based decision engine and method, and relates to the technical field of financial risk control, and the decision engine comprises a strategy warehouse module, a mapping service module, an execution engine module, a data management module, a version rollback module, a security scale module, and a monitoring alarm module. According to the invention, function decoupling is realized through collaborative design of the seven modules, and the modules interact through a unified interface, support strategy dynamic loading, gray release, distributed parallel execution, version governance and security compliance capabilities, and ensure high cohesion and low coupling of the system; through unique strategy version management, intelligent mapping and gray release, plug-in parallel execution, multi-version consistency view, block chain audit rollback, security compliance management and control, intelligent monitoring and other technical innovations, a complex, dynamic and high-concurrency risk control business scene is effectively supported; the security, the flexibility, the expandability and the auditing performance of the decision engine are remarkably improved, and the technical advancement and the industry creativity are high.