357 results about "Computational logic" patented technology

The invention relates to the technical field of data information processing, in particular to an artificial intelligence-based index layer model construction analysis method and system, and the method comprises the following steps: 1, multi-source data synchronization and quality evaluation; 2, service context analysis and subject domain intelligent division are carried out, and service document and data table field semantics are analyzed; 3, index demand intelligent analysis and calculation logic generation: analyzing index demands input by a user through a natural language processing technology, matching service entities in the knowledge graph, generating initial calculation logic and optimizing execution efficiency; and 4, dynamically updating the index data and verifying the consistency, executing the optimized computational logic to generate the index data, tracking the blood relationship of the data, triggering dynamic updating, and setting a threshold value according to a historical fluctuation range to carry out abnormal alarm. Through automatic and intelligent processing, the probability of human intervention and human errors is reduced, and the efficiency and accuracy of data processing are improved.

The invention relates to the technical field of network communication, and discloses a vehicle-mounted edge computing data recording system and method based on protocol adaptive analysis, and the system comprises a multi-protocol network access controller which is used for capturing an original data frame and extracting a physical feature triple; the identification analysis engine is used for executing hash operation on the triple to generate a physical feature code and retrieving a physical logic address mapping table; the direct memory access controller is used for responding to a target memory address pointer hit by retrieval and directly writing a data load into an input buffer area of the functional operation module, and by constructing a direct addressing mechanism based on Hash mapping, thorough decoupling of vehicle-mounted heterogeneous network physical topology and edge computing logic is achieved.

The invention provides an intelligent question number and index management engine and system based on dynamic reward optimization, relates to the technical field of data learning, and is used for enterprise index analysis, attribution diagnosis and decision support. The system is provided with an index governance layer, index caliber, computational logic, blood relationship, credibility score and version information are managed in a unified mode through a dynamic knowledge graph, unified semantic constraint is carried out on a multi-agent analysis process, and index consistency and traceability are guaranteed. The system also establishes a causal cognition module, based on time sequence data and in combination with expert priori, generates and corrects a business index causal directed acyclic graph, realizes root cause positioning and anti-fact simulation, and answers what change is and what intervention is. The multi-agent collaborative analysis core is responsible for natural language intention analysis, index compliance verification, automatic access, causal inference, narrative generation and chart presentation, calculates a multi-target composite reward value based on user feedback and interaction behaviors, and adaptively adjusts output; and the user corrects and writes back to form closed-loop learning.

The invention discloses a human resource data natural language query and SQL generation method and system, and relates to the technical field of SQL generation, and the method comprises the steps: preprocessing a natural language query text input by a user, and constructing a query semantic tree; traversing the query semantic tree to generate a candidate entity set, determining an unambiguous entity through a three-layer ambiguity resolution mechanism of semantic scoring, service verification and cross validation, identifying a composite intention, disassembling the composite intention into a sub-intention sequence, and generating an SQL operation instruction set in combination with a preset SQL operation mapping rule; based on the SQL operation instruction set and the human resource data association model, determining a data table set and an association path, generating a dynamic calculation logic and structured condition expression, and constructing a structured query skeleton; according to the method, the personalized SQL template is dynamically generated according to the query skeleton, after filling, grammar verification, strategy optimization and logic consistency verification are performed, and the optimized SQL statement is output, so that the query accuracy and performance are remarkably improved.

The invention relates to the technical field of railway contact networks, and provides a railway contact network icing thickness prediction method and system. The method comprises the following steps: inputting meteorological environment parameters into a machine learning model, and carrying out dynamic calibration on icing medium parameters based on the meteorological environment parameters by utilizing the machine learning model to obtain calibrated icing medium parameters; obtaining an icing growth coefficient based on the meteorological environment parameters, the conductor physical characteristic parameters and the calibrated icing medium parameters; and inputting the calibrated icing medium parameters and the icing growth coefficient into a Makkonen model to obtain an icing thickness prediction value. According to the technical scheme provided by the invention, the key icing parameters are dynamically calibrated through machine learning, and the computational logic of the Makkonen model is optimized, so that the icing thickness prediction precision under complex meteorological conditions is improved, and the problem of prediction failure caused by insufficient environmental adaptability of a traditional static model is solved.

The invention belongs to the field of data processing, particularly relates to a standardized processing method, system and equipment for original engineering parameters of a flight simulator, and aims to solve the problems of serious field limitation, low processing efficiency and poor adaptive capacity in the prior art. The method comprises the following steps: analyzing a predefined configuration file, and dynamically generating a database temporary table matched with an original data file structure according to field structure information and a parameter mapping rule defined in the configuration file; directly loading original engineering parameter files into the temporary table by adopting a batch loading mechanism based on a file system path, and executing multi-file parallel import based on a task queue; and converting calculation logic in the parameter mapping rule into an executable query expression, and directly exporting a standardized parameter file after calculation is executed through the temporary table. According to the method, field limitation is broken through, efficient processing of mass data is achieved, instant updating of service logic is achieved through configuration rules, and the maintenance cost is greatly reduced.

The invention relates to the technical field of natural language processing, and discloses a generated text quality processing method and system based on a large language model, and the method comprises the steps: constructing a logic topological graph and a Laplacian matrix of an original generated text, and extracting a feature value sequence; recognizing a logic bearing wall based on the attention gradient and generating an anti-fact contrast text; calculating a logic collapse index by combining the difference between the map characteristic values of the original text and the contrast text and the logic polarity overturning condition; and judging the quality of the generated text according to the logic collapse index, blocking the text which does not pass the judgment, and shaping and resampling the output probability value of the error position by utilizing the spectrum difference information to generate a new text. The method does not need to depend on an external knowledge base, quantifies the stability of text logic through anti-fact interference and spectrum analysis, and identifies a high-risk illusion text; and a wrong logic path is automatically corrected through Logits shaping, so that the logic self-consistency of the generated content is improved on the premise of ensuring the semantic smoothness of the text language.

The embodiment of the invention discloses a method and device for constructing a field consanguinity tree, a storage medium and a terminal. Firstly, a target field, a database query statement of the target field and upstream blood relationship information are obtained, the target field serves as a root node, a first executable calculation function packaging calculation logic of the target field is generated, and the calculation process is dominant and modularized. By analyzing input parameters of the function, an upstream field which directly depends on is automatically identified and determined to serve as a child node. And then recursively processing each sub-node, generating a corresponding calculation function, analyzing the dependency of the calculation function, tracing layer by layer until all leaf nodes (namely original bottom table fields), outputting a complete blood relationship tree containing all the nodes and the corresponding calculation function, and displaying the complete blood relationship tree through a visual interface. According to the method, automatic and accurate tracing from a target field to an original data source is realized, a computable and reusable blood relationship knowledge framework is constructed, and a solid foundation is provided for data understanding, problem investigation and intelligent data service.

The invention discloses an injection-production regulation imbalance quantitative diagnosis method based on fluid potential energy, and the method comprises the steps: extracting a streamline track space coordinate sequence and corresponding point oil reservoir parameters based on a dynamic permeability field, a pressure field and an injection-production streamline field; constructing a three-dimensional fluid potential energy field model in the injection-production unit, and dividing dominant seepage, a retention zone and a transition zone based on the injection-production unit through potential energy gradient tensor; establishing a seepage potential index LS by taking the flow line penetrating pore volume as a stability weighting factor and combining the potential energy gradient intensity; on the basis of LS sorting, a potential energy-volume two-factor accumulation curve is constructed, the development non-equilibrium degree is represented through curve form parameterization, and the envelope area of the accumulation curve is calculated and used for representing the oil reservoir flow field development non-equilibrium degree and the development effect of each type of injection-production unit. According to the method, fluid potential energy and a two-factor cumulative distribution concept are introduced into computational logic, and unified and quantitative characterization of an oil reservoir flow field and an injection-production unit is realized in combination with a seepage mechanism of the injection-production unit.

The embodiment of the invention discloses a method and device for analyzing a data index, equipment and a medium, and the scheme can comprise the steps: generating a first graph structure used for representing the computational logic of a first data index based on the computational logic code of the first data index; generating a second graph structure for representing the computational logic of a second data index based on the computational logic code of the second data index; performing difference judgment on the first graph structure and the second graph structure to obtain a judgment result; and according to the judgment result, generating description information for representing statistical aperture difference information of the first data index and the second data index.

The present application provides a TVM-based NPU neural network model deployment method and device, including: obtaining the neural network model to be deployed and the hardware attribute information of the current NPU; determining the convolution operator to be deployed to the current NPU based on the computational logic of the neural network model; determining the optimal parameter configuration of each convolution operator in the current NPU based on the hardware attribute information of the current NPU and the parameter information of each convolution operator; adjusting the data layout and convolution weight of the TIR layer and modifying the parameters of the relay layer based on the optimal parameter configuration and the hardware attribute information of the current NPU to obtain a target TVM adapted to the current NPU; generating a deployment file for the neural network model suitable for the current NPU based on the target TVM. By adjusting the parameters of the TVM's TIR layer and relay layer accordingly, the obtained target TVM can be adapted to the NPU.

The invention relates to an anti-reverse confusion method and system for an end-side equipment deep learning model, and the core process comprises the steps: firstly inputting an original model obtained through the training of a model training frame into a deep learning compiling frame, and extracting three types of key information, namely, a model operator, a topological structure, parameters and dimensions, through the characteristics of a compiler; then constructing a feature analysis module to evaluate model features, dynamically matching a confusion scheme from a strategy library, and balancing safety and performance; the confusion module is embedded into a plurality of different levels such as a computational graph level, an operator template level and tensor intermediate expression through a hierarchical compiling mechanism, and a complex scheme can be jointly implemented across multiple levels; and finally, the compiler synchronously completes confusion reinforcement when generating the target code. According to the method, hardware adaptation is not needed, low-overhead confusion is achieved through a native pass mechanism of a compiler, fine-grained customized protection is supported, a model structure, parameters and computational logic can be effectively hidden, reverse engineering attacks can be resisted, and the method is particularly suitable for end-side equipment scenes with limited computing power.

The invention discloses a method for constructing a Text2MDX framework based on an intelligent agent, and relates to the technical field of natural language processing, and the method comprises the steps of constructing an intelligent agent architecture, constructing a chain structure composed of a semantic analysis intelligent agent, a rule memory intelligent agent and an MDX synthesis intelligent agent, and constructing a Text2MDX framework. The method has the advantages that the chain structure composed of the semantic analysis intelligent agent, the rule memory intelligent agent and the MDX synthesis intelligent agent is constructed, the semantic analysis intelligent agent can accurately disassemble the text and extract potential self-defined semantic rules in multi-round dialogue or long text processing, in the whole process, the intelligent agent chains cooperate in order, and the user-defined semantic rules are extracted; the cross-sentence context logic is automatically maintained, the situation that the MDX expression lacks key calculation logic is avoided, the accuracy and reliability of conversion from the natural language to the MDX expression are remarkably improved, more accurate data analysis support is provided for a user, and the problem of context breakage of a complex interaction scene is solved.

The invention discloses a reusable key value cache generation method and device, equipment, a storage medium and a product, and relates to the technical field of computers.The reusable key value cache generation method comprises the steps that an obtained text block is input into a preset large model, and a reusable key value cache corresponding to the text block is output based on a preset fixed prefix key value, the preset large model is obtained by optimizing calculation logic of an attention layer of the original large model based on an attention distribution mechanism. According to the method, the reusability of the key value cache is realized through the preset large model based on the preset fixed prefix key value, the problem of unstable model output caused by position coding and attention distribution changes is effectively solved, and the reuse efficiency of the key value cache is remarkably improved. Meanwhile, the computational logic of the attention layer of the model is optimized, so that the performance and stability of the model during multiplexing of the key value cache of the text block are improved.

The invention relates to the technical field of derivative variable calculation, and discloses a derivative variable calculation method and system based on a graph algorithm, and the method comprises the steps: analyzing derivative variable definition and calculation logic through employing a variable SQL syntax tree analysis technology, taking an analyzed variable as a node, determining a variable dependency relationship as an edge according to a processing task, and carrying out the calculation of the derivative variable; constructing a directed acyclic graph and storing the directed acyclic graph in a graph database; the nodes store variable metadata, execution priorities and calculation information, and store calculation task metadata and calculation information; according to node priorities, finding out high-priority variable nodes, querying all in-degree edges of the high-priority variable nodes, forming one or more subtrees, forming high-priority execution subtrees, performing topological sorting on the execution subtrees, and performing execution by an execution engine; the method aims to solve the problems of low calculation efficiency and insufficient utilization of calculation resources in the prior art.

The invention discloses a knowledge graph inference method based on prior knowledge enhancement, and mainly solves the problem of inaccurate reward caused by insufficient knowledge graph prior knowledge in the existing knowledge graph inference method based on reinforcement learning, thereby optimizing a model more effectively. The method comprises the following steps: reasoning path searching based on reinforcement learning; calculating an answer correctness reward; logic rationality reward calculation; a reward enhancement strategy based on path importance; performing logic rationality reward enhancement based on large language model context learning; and optimizing the model. According to the method, on the basis of an existing knowledge graph reasoning method based on reinforcement learning, huge internal knowledge of a large language model is fused into reward calculation in an efficient and information loss resistant mode, and the problem that rewards are inaccurate due to insufficient prior knowledge of a knowledge graph is solved. The performance of the knowledge graph reasoning method based on priori knowledge enhancement is remarkably improved compared with that of an existing method.

The application provides a smart question and index management engine and system based on dynamic reward optimization, relates to the technical field of data learning, and is used for enterprise index analysis, attribution diagnosis and decision support. The system sets an index management layer to uniformly manage index caliber, calculation logic, blood relationship, credibility score and version information by means of a dynamic knowledge graph, uniformly implements semantic constraints on a multi-agent analysis process, and guarantees index consistency and traceability. The system also establishes a cause-effect cognition module, generates and corrects a business index cause-effect directed acyclic graph based on time series data and in combination with expert priori, realizes root cause positioning and counterfactual simulation, and answers "why change" and "what if intervention". A multi-agent collaborative analysis core is responsible for natural language intention analysis, index compliance verification, automatic data acquisition, cause-effect inference, narrative generation and chart presentation, and calculates a multi-target compound reward value based on user feedback and interaction behavior, and adaptively adjusts output; user correction write-back forms a closed-loop learning.

The invention discloses a natural language processing system encryption method, electronic equipment and a computer readable medium. The method comprises the following steps: acquiring a character mapping relationship between an original character and an encrypted character determined based on a character encryption algorithm; according to a Token ID transformation algorithm, a Token ID mapping relation between the original Token ID and the transformed Token ID is determined; and according to the character mapping relationship and the Token ID mapping relationship, encrypting a pre-trained natural language processing system, so that the encrypted natural language processing system can output a ciphertext response based on an input ciphertext prompt word. Side channel attacks and middle eavesdropping can be resisted, the exposure risk of original data is remarkably reduced, the data security is improved, mathematical reconstruction of computational logic of the system is avoided, so that the computational overhead is remarkably reduced, the reasoning delay is reduced, the computational efficiency and the response real-time performance are improved, a natural language processing system does not need to be retrained, and the method is easy to implement. The calculation overhead is further reduced, and the encryption efficiency is improved.

The invention provides a sensitive data collaborative auditing platform and method based on multi-party network security computing, and relates to the technical field of multi-party security computing. The platform comprises a data input layer which is used for each data holder to acquire original data, standardize the original data, perform differential privacy processing on the standardized data, encrypt private data by adopting a Paillier homomorphic encryption algorithm to obtain a ciphertext, segment the ciphertext by adopting Shamir secret sharing to obtain a plurality of segment ciphertexts, uploading the fragmented ciphertext to a confusion layer; the confusion calculation layer is used for aggregating the fragmented ciphertexts to obtain aggregated ciphertexts, fragmenting the aggregated ciphertexts by adopting Shamir secret sharing to obtain a plurality of fragmented aggregated ciphertexts, and distributing the fragmented aggregated ciphertexts to audit layer nodes; performing target calculation on the fragmented aggregation ciphertext by adopting MPC calculation logic; and the auditing execution layer is used for converting the auditing rule into MPC computational logic. According to the application, the security, efficiency and cross-domain compatibility of multi-party collaborative auditing can be improved.

The invention provides a data processing method and device, electronic equipment and a storage medium. And selecting the target script file according to the type of the rule configuration parameter. The target script file is a universal script template file corresponding to the type. The rule configuration parameters are generated into the target rule file based on the format of the target script file, the rule instance existing in the script template file can be replaced, and the new computational logic and the rule configuration parameters in the target rule instance are compiled into the target script file in real time. And the target rule file is updated and stored in a rule file library in a broadcast mode, namely, the target rule file is loaded to the rule file library. The target rule file can be directly and dynamically compiled and loaded, so that the target rule file becomes a pluggable logic unit, the risk management system can realize non-perceptual online of the risk assessment rule without stopping running, and the processing timeliness of the risk control business can be improved.

The invention discloses a risk control rule generation method and device based on a large model and computer equipment. The method comprises the steps of obtaining a to-be-analyzed law and regulation policy entry; automatically analyzing to-be-analyzed laws and regulations and policies by using the large model, and identifying and extracting review structure information; converting the examination structure information into a structured rule expression, and defining a corresponding quantitative index name and computational logic; checking the data index completeness; when the verification is passed, a Python code is automatically generated; checking the Python code by adopting a dual mechanism of static grammar check and dynamic sandbox test; integrating and storing the review structure information, the generation result and the Python code passing the verification in a risk review rule base; and when the new review object is reviewed, obtaining the index data, matching the applicable rule, executing the corresponding code, and summarizing the review result. By implementing the method provided by the invention, deep fusion of policies and regulations can be realized, and efficient generation and dynamic update of risk control rules can be realized.

The invention relates to the technical field of artificial intelligence chips, and provides an operator performance evaluation method and device, equipment, a storage medium and a program product, and the method comprises the steps: obtaining a configuration file corresponding to a to-be-evaluated operator, enabling the configuration file to comprise a hardware configuration file and an operator configuration file, carrying out the instantiation of a global object based on the configuration file, obtaining a hardware platform instance and an operator instance; the operator instance inherits a base class in a unified manner, and the base class is a standardized interface obtained by abstracting computational logic of performance indexes of operators; the global object is a class object obtained by performing interface abstraction on a hardware platform interface and an operator interface; according to the performance index of the base class calculation operator instance when the base class calculation operator instance runs on the hardware platform instance, the evaluation result of the theoretical performance of the to-be-evaluated operator is obtained, multiplexing of the calculation logic of the operator performance index is achieved, the calculation logic of the operator performance index does not need to be rewritten for different hardware platforms, and the calculation efficiency is improved. And the evaluation efficiency of operators across hardware platforms is improved.

The invention provides an operator compiling method and device, equipment and a medium, and relates to the technical field of artificial intelligence. The operator compiling method comprises the following steps: obtaining a to-be-compiled operator source code, wherein the operator source code is a code compiled by adopting a field-specific language based on Triton; converting the operator source code into Triton intermediate representation, wherein the Triton intermediate representation is used for representing computational logic of the operator source code; carrying out optimization which does not involve chip information on an operator based on the Triton intermediate representation to obtain an optimized intermediate representation; and compiling the optimized intermediate representation through a Triton compiler corresponding to the target chip to obtain a target code. Open sources can be conveniently realized in partial stages in the operator compiling process, so that the partial compiling process of the operator can be conveniently optimized and expanded based on the open sources.

A computing system is provided, including one or more processing devices configured to receive an extended stabilizer form of a quantum error correction code. The one or more processing devices are further configured to receive a logical Clifford operation specification of a logical Clifford operation. Based at least in part on the extended stabilizer form, the one or more processing devices are further configured to compute a stabilizer tableau of the quantum error correction code. Based at least in part on the stabilizer tableau and the logical Clifford operation specification, the one or more processing devices are further configured to compute a physical representation matrix of the logical Clifford operation. The one or more processing devices are further configured to output the physical representation matrix.

The invention discloses a privacy protection data classification method and system based on a cloud platform, and relates to the field of data classification management, and the method comprises the steps: distributing sensitivity tags for different data attributes at a client according to a preset strategy, and carrying out the encryption through selecting a homomorphic encryption public key matched with the tag level, and forming a ciphertext beam containing the ciphertext and the label. And then, at the cloud, guiding a subsequent calculation process by using the tag vector in the ciphertext beam. Through a self-adaptive encryption model, different calculation paths can be dynamically selected to execute homomorphic feature engineering and classification calculation according to whether high-sensitivity data exists in a label or not. According to the method, refinement and isolation processing of data with different sensitivities is realized, an encryption strategy and calculation logic are deeply bound, and dynamic balance between calculation efficiency and security level is realized on the premise of ensuring data security.

The invention discloses a bid invitation file compiling method and system based on artificial intelligence, a medium and a product, and relates to the field of intelligent contract processing. The method comprises the steps that a file compiling system divides a natural language text into first text content containing computational logic and common second text content through semantic analysis; then, the first text content is converted into an executable formalized rule expression, and an association mapping relation between the executable formalized rule expression and the first text content is established; and calculating and generating a derivation record containing parameters, calculation steps and dependency relationships according to the expression. And finally, embedding a unique tracing identifier in the generated clause text, and querying and displaying a derivation record through the association mapping relationship by triggering the identifier by a user. According to the method, a complete traceable link from the natural language to the calculation result is realized, so that auditing personnel can clearly understand the logic source of the calculation result, the interpretability and the credibility of the generated calculation terms are improved, and the verification efficiency and the accuracy of the bid invitation file are improved.

The invention provides a lightweight detection method for a small target of an intelligent unmanned vehicle, and belongs to the field of target detection in an artificial intelligence driving scene, YOLOv5s is taken as a reference model, and the specific improvement comprises the following steps: embedding an efficient multi-scale attention (EMA) mechanism behind P3 / 8 and P4 / 16 feature layers of a backbone network, and enhancing the feature perception ability of the model to the small target; a check network is improved, common convolution is replaced by a Ghost module, a C3 module is replaced by a C3Ghost module, and model parameters and calculation amount are reduced to realize light weight; and proposing an SU-WIoU loss function, introducing positioning uncertainty parameter correction anomaly calculation logic, and improving the regression precision of the bounding box of the small target. According to the invention, through collaborative improvement of backbone, check and loss functions, the conflict defect of the existing method on the aspects of small target detection precision and lightweight adaptability is overcome.

The invention discloses an anomaly monitoring method for substation equipment and related equipment, relates to the technical field of anomaly monitoring of a power system, and solves the problem of large misjudgment rate difference during anomaly monitoring of different types of substation equipment in the prior art. According to the method, feature weight distribution is independently performed on various types of substation equipment, so that the spectral clustering algorithm can extract key features of each type of equipment and perform clustering based on the key features, and misjudgment caused by feature distribution difference is avoided; and in combination with unified computational logic, the cross-device type consistency of the abnormal monitoring result can be ensured. By adopting the embodiment provided by the invention, the abnormal conditions of different types of substation equipment can be accurately detected.

The invention discloses a code running time determination method based on a CORTEX M4 chip and related equipment. The method comprises the following steps: establishing a C + + operation support environment for a CORTEX M4 chip; defining a time measurement class, taking system time recorded by a constructor function of the class as starting time, and taking system time recorded by a disruptor function of the class as ending time; and determining the running time of the target code block based on the starting time and the ending time. The problems that at present, a system beat interface is called before and after a code segment to obtain a timestamp, difference is made to measure the running time of the code segment, and repeated time obtaining and difference calculation logic needs to be inserted into each code segment needing to be measured, so that code redundancy and structure bloating are caused can be solved.