43 results about "Algorithmic logic" patented technology

A conversational AI platform for managing remote and hybrid clinical trials for investigational medical interventions. Embodiments of the present disclosure comprise a conversational AI model and an algorithmic logic engine configured to define and implement protocol parameters for dosing schemas, visit schedules, safety thresholds, and eligibility criteria for a clinical trial. An AI agent is configured to deliver mapped voice prompts to participant devices, captures audio responses, transcribe and extract symptom, adherence, or adverse-event data, and pair response data with physiological-sensor or laboratory input data. The logic engine continuously evaluates the combined data to adaptively select dose-escalation or titration instructions and issue follow-up queries while enforcing safety thresholds. All prompts, audio, transcriptions, decisions, and metadata may be immutably timestamped in an electronic record repository accessible via role-based, encrypted connections. Embodiments of the present disclosure provide for real-time, audit-ready trial communications, automated personalized dosing, and enhanced participant safety monitoring.

The invention relates to the technical field of electric power security, and discloses a regional building group electric power transaction security guarantee system, which comprises the following modules: a trusted computing environment authentication module, a transaction data encryption and commitment module, an under-chain privacy computing and verification module, and a computing process certification and settlement module. According to the method, the algorithm hash value recorded in the program registry contract is combined with the remote authentication mechanism of the trusted execution environment, so that the traceability and verifiability of the under-chain computing environment are constructed. The system digitally signs a report containing an algorithm code hash value by using a hardware built-in private key, and submits the report to an on-chain contract for comparison, and the mechanism ensures that any under-chain privacy computing task is executed in a specified algorithm environment which is approved by block chain consensus and is not tampered with codes, so that the privacy computing efficiency is improved. The security risk caused by the untrusted computing environment or the non-compliance of the algorithm logic is completely eradicated from the source, and the basic guarantee is provided for the correctness of the subsequent computing result.

The present application relates to the field of artificial intelligence, and provides a multi-chip operator automatic generation method, device and equipment, the method comprising: obtaining operator demand description; in response to the operator demand description, generating a standard operator code conforming to a preset intermediate representation specification and independent of specific hardware; based on a hardware rule library of a target chip, converting the standard operator code into target operator code adapted to the target chip; wherein the hardware rule library at least includes application programming interface mapping rules, hardware characteristic constraint rules and performance and engineering experience rules; compiling and verifying the target operator code, and outputting the target operator code that passes the verification. The present application solves the problem of poor reusability and high migration cost of operator automatic generation logic in the prior art in a multi-chip scenario, realizes decoupling of algorithm logic generation and chip adaptation, and efficient and controllable multi-platform operator automatic generation.

This information processing device: acquires corporate enterprise designation information that designates a first corporate enterprise to which leader information, which pertains to a substantive leader, is to be provided; acquires corporate enterprise information having beneficiary-related information that includes at least beneficiary information representing a beneficiary of the subject corporate enterprise and ownership ratio information representing the ratio at which the beneficiary owns stocks of the subject corporate enterprise; refers to the acquired corporate enterprise information and specifies the beneficial ownership rate according to a definition in an assumed designated nation; specifies a beneficiary who is the substantive beneficiary in the first corporate enterprise, on the basis of the acquired corporate enterprise information and the specified beneficial ownership rate; and extracts at least some of the beneficiary-related information corresponding to the specified beneficiary as leader information. The device further comprises a rule-set storage unit storing a plurality of rule-sets each defining a complete algorithmic logic for specifying the beneficiary according to a respective jurisdictional definition, a jurisdictional routing unit for selecting rule-sets for execution, a multi-result publication unit for comparative output of specification results from multiple rule-sets, and a scoring unit for calculating descriptive and qualitative scores characterizing the complexity and structural aspects of each beneficiary specification process.

The invention discloses a cryptographic intelligent question and answer method based on maximum marginal correlation retrieval, and relates to the technical field of natural language process.The method comprises the steps that firstly, a cryptographic field document is subjected to analysis and semantic partitioning, and the integrity of mathematical formulas and algorithm logic is ensured by dynamically calculating text similarity; and information splitting caused by traditional fixed partitioning is avoided. And then, obtaining knowledge fragments which are related to the question and have various contents from the vector database by utilizing a maximum marginal correlation retrieval algorithm, and providing multi-angle reliable contexts for answer generation. And finally, through field adaptive prompt engineering and dynamic parameter adjustment, guiding the large language model to play a role of a cryptography expert based on the retrieved context, and generating an accurate personalized answer meeting the user demand.

The invention discloses a data circulation transaction method and device, a storage medium and electronic equipment, and relates to the technical field of data processing, and the method comprises the steps: carrying out the reasoning of an algorithm execution element generated based on a business target through an algorithm agent, and obtaining an algorithm implementation scheme; the algorithm implementation scheme comprises an algorithm logic description and a code adapted to an execution environment; performing data calculation on the algorithm implementation scheme by using a data product deployed at the data service node to obtain a data calculation result; and when it is determined that the data calculation result meets the transaction confirmation conditions of the two transaction parties, the data routing node is utilized to send tradable information to the data demand side, so that the two transaction parties achieve data transaction confirmation. The data transaction risk can be reduced, and the supply-demand matching efficiency is improved.

Examples disclosed herein may estimate locations of players not visible in a sporting broadcast video. A prediction model may be generated based on a training data set of in-venue tracking data that includes locations of all players at all times and the corresponding broadcast tracking data that may not necessarily contain the locations of all players at all times. The prediction model may be based on an algorithmic logic (e.g., a spline regression) or machine learning model (e.g., k-nearest neighbor, deep neural network). The generated predicted model may be used to estimate the unknown locations of players in broadcast tracking based on the known locations.

The application relates to the technical field of radar signal processing, in particular to a design method of a software-based radar data processing component based on a VSIPL standard, which comprises the following steps: constructing a layered component model composed of a support library layer, an algorithm logic layer and a memory space management layer; performing algorithm-level granularity division on a radar data processing flow, and decomposing the entire radar data processing flow into multiple algorithm-level modules; designing API interfaces of each function based on the layered component model; using a deep performance optimization strategy based on a VSIPL middleware, and uniformly replacing function functions frequently called in the radar data processing flow into VSIPL standard middleware optimized by a bottom-layer instruction set; and based on the multiple algorithm-level modules and actual reuse requirements, performing secondary aggregation, and encapsulating algorithm-level modules that can be reused into composite components. The method improves the efficiency and real-time performance of radar data processing, enhances the flexibility and expandability of the radar data processing component, and reduces development and maintenance costs.

Apparatuses and methods for managing a coherent memory are described. These may include one or more algorithmic logic units (ALUs) and an input / output (IO) interface. The I / O interface may receive one or more commands and retrieve data from or write data to a memory device. Each command may contain a memory address portion associated with a memory device. The apparatus may also include a memory mapping unit and a device controller. The memory mapping unit may map the memory address to a memory portion of the memory device, and the device controller may communicate with the memory device to retrieve data from or write data to the memory device. The apparatus may be implemented as a processing element in a configurable logic block network, which may additionally include a control logic unit that receives programming instructions from an application and generate the one or more commands based on the instructions.

The invention provides a function upgrading method, device and system applied to the Internet of Things and electronic equipment. According to the method and the device, the upgrading information related to the to-be-upgraded function is obtained from the algorithm platform, and the target algorithm upgrading package identifier required to be used during function upgrading is determined from the plurality of algorithm upgrading package identifiers carried in the upgrading information, so that the target algorithm upgrading package is upgraded under the condition of determining that the residual resources in the Internet of Things equipment can accommodate the target algorithm upgrading package; and downloading the corresponding target algorithm upgrade package to perform function upgrade on the Internet of Things equipment. According to the method, the technical concept that different algorithm logics related to functions are uniformly packaged into a whole package and upgrading is carried out in a whole package mode is broken through, the algorithm functions are associated with the multiple algorithm logics, only the algorithm package corresponding to the algorithm logic needing to be upgraded needs to be independently upgraded during upgrading, and the upgrading efficiency is improved. The problems that the equipment end side occupies a large amount of resources and the downloading time is long when the function of the Internet of Things equipment is upgraded are solved.

The invention relates to a method for realizing an LD (Load) assembly instruction of a special processor chip based on an LLVM (Load-Load) rear end, and aims at a special architecture of an LD instruction in an instruction set of the special processor chip, a set of complete mapping mechanism from an intermediate representation to a target machine instruction is constructed by newly establishing a rear end processing flow of an LLVM compiling framework, and the LD (Load) assembly instruction of the special processor chip is obtained. And conversion from a high-level language to a chip special assembly instruction is realized. In the legalization stage of the rear end of the LLVM, unified architecture adaptation processing is carried out on global address access and a vector / scalar loading mode through a user-defined ISelLowing mechanism and instruction selection. According to the method, the intermediate language representation is converted into the target machine instruction for the special processor chip by utilizing the LLVM rear end, and according to the method, a developer can directly realize algorithm logic in a high-level language stage without manually writing a bottom assembly code, so that the development complexity is reduced, and the portability and maintainability of the code are improved.

The invention provides a workflow execution method applied to industrial control and a related device. According to the invention, a workflow arrangement architecture can be adopted, and an industrial control algorithm is converted from a hard coding mode to a visual and modularized workflow arrangement mode, so that a user can arrange the industrial control workflow in a visual mode. Under the condition that the industrial control workflow (namely the target workflow) is obtained, the workflow configuration information of the workflow can be loaded, the target node execution graph is constructed according to the workflow configuration information, and the node execution logic of the corresponding control node is loaded and operated according to the target node execution graph, so that the industrial control process can be orderly propelled according to the preset sequence. Therefore, hard coding can be replaced by nodal arrangement, and dynamic execution of the algorithm logic can be realized, so that the algorithm logic of the industrial control algorithm can be quickly adjusted and recombined, and the flexibility of the industrial control process can be improved.

The invention discloses an algorithm model management system supporting model configuration iteration, and relates to the technical field of algorithm model development management, the algorithm model management system is composed of a configuration layer, an execution layer, a data layer and a monitoring layer, the configuration layer is used for defining and managing all configuration information required by model iteration; the execution layer is used for automatically executing a complete process from data preparation to model online according to an instruction of the configuration center; the data layer is used for managing model assets and versions; and the monitoring layer is used for monitoring online model performance in real time and automatically triggering a model iteration process according to a preset rule. According to the method, an algorithm engineer can concentrate on algorithm logic through a configuration driving mode, and does not need to care about complex engineering implementation and operation and maintenance details, so that rapid and automatic iteration and delivery of an algorithm model are realized.

The invention relates to the technical field of logistics storage and industrial automation, in particular to a package production line intelligent stack type generation method based on heuristic knowledge guidance, and the method comprises the steps: obtaining initial parameter data, building a coordinate system according to the initial parameter data, generating an original stack type solution set based on initial parameters and the coordinate system, and generating a package production line stack type solution set; performing stack type fine tuning processing on the original stack type solution set, performing similar stack type reduction on the regularized stack type solution set, and performing overturning decision according to the reduction solution set to screen an optimal scheme, including counting the number of overlapped boxes before and after overturning so as to divide stability levels, sorting stack types in combination with the number of boxes, and outputting an optimal stack type scheme set; the optimal stack type scheme set is verified, and a final result is output, the method can flexibly adapt to combinations of trays and boxes of different specifications, algorithm logic does not need to be developed again for specific sizes, and the problem that a traditional method is poor in adaptability under the scenes of multi-size box bodies and multi-specification trays is solved.

The invention relates to the field of building information model construction, and discloses a building rectangular structural column reinforcement cage parameterized modeling method, which comprises the steps of building a building information model, initializing a project environment through a user interface, reading geometric attributes and non-geometric attributes of a structural column model, and building a rectangular structural column reinforcement cage. Reinforcing bar information in the non-geometric attributes is converted into structured reinforcing bar data, a longitudinal bar model is established on the surface of the structural column according to the plane coordinates, the three-dimensional coordinates of binding points of each stirrup on the longitudinal bars are calculated at the same time according to a special-shaped stirrup circulating winding method, three-dimensional modeling of the stirrups is carried out according to the coordinates of the binding points, and the three-dimensional modeling of the stirrups is carried out. Building of a building rectangular structure column steel reinforcement framework model is completed; according to the method, the stirrup binding points are automatically calculated on the basis of the longitudinal bar coordinates by adopting a special-shaped stirrup circulating winding method, automatic processing is carried out by directly reading geometric attributes and non-geometric attributes of the model, the problems of manual intervention and data synchronization are solved, complex design rules are converted into algorithm logic, and intelligent and compliant modeling of the whole process is achieved.

The invention provides a rail transit multi-device task scheduling method and device, and relates to the technical field of computers. According to the method, a device data processing flow is abstracted into a reusable computing topology template, and thorough decoupling of algorithm logic, specific devices and runtime context is realized, so that the same set of processing flow can be reused by all devices of the same type, and the system configuration efficiency and maintainability are greatly improved; furthermore, by means of an automatic instantiation and topology scheduling mechanism, the system can dynamically construct an executable task graph according to event triggering, and perform parallel scheduling based on a dependency relationship, so that the complexity of task arrangement and execution is remarkably reduced, the strategy deployment can be scaled, and the expandability of the system is higher.

The invention belongs to the technical field of integrated circuit design and information security, and particularly relates to a fault detection method and system for a cryptographic chip. The fault detection method for the cryptographic chip comprises the following steps: when the cryptographic chip runs a cryptographic algorithm, injecting a preset test vector into input data of algorithm logic; obtaining an actual output result of the algorithm logic corresponding to the test vector; comparing the actual output result with a pre-calculated expected output result under a fault-free condition; and judging whether the password chip has a fault or not according to a comparison result. The invention also correspondingly provides a system for realizing the method. According to the method, the test vector is actively injected and the output result is compared, so that the calculation error of the cryptographic chip caused by hardware faults, side channel attacks or natural aging and the like can be effectively detected, and the method has the advantages of high detection real-time performance, low hardware overhead, high universality and the like.

The invention discloses an adaptive neighborhood point cloud fusion method based on voxel pre-acceleration, and belongs to the technical field of three-dimensional data processing, and the method comprises the following steps: S1, collecting three-dimensional point clouds, and forming a total point cloud; s2, performing space division on the total point cloud to generate a Hash dictionary; s3, traversing the Hash dictionary to generate a neighborhood point cluster; and S4, calculating representative points for each neighborhood point cluster, and taking a set of all the representative points as a final fusion point cloud. The algorithm logic is clear, engineering implementation, deployment and parallelization acceleration are easy, and the method has high robustness and engineering value in practical application.

The application relates to the field of building information model construction, and discloses a building rectangular structure column reinforcement framework parameterized modeling method, which comprises the following steps: establishing a building information model, initializing a project environment through a user interface, reading geometric properties and non-geometric properties of a structure column model, converting reinforcement information in the non-geometric properties into structured reinforcement data, establishing a longitudinal reinforcement model on the basis of the plane coordinates on the structure column surface, simultaneously calculating three-dimensional coordinates of a binding point of each stirrup on the longitudinal reinforcement according to a special-shaped stirrup cyclic winding method, and performing three-dimensional modeling of the stirrup according to the binding point coordinates to complete creation of a building rectangular structure column reinforcement framework model; the special-shaped stirrup cyclic winding method is adopted to automatically calculate the binding point of the stirrup based on the longitudinal reinforcement coordinates, and automatic processing is performed by directly reading the geometric properties and the non-geometric properties of the model, so that manual intervention and data synchronization problems are eliminated, complex design rules are converted into algorithm logic, and intelligent and compliance modeling of the whole process is realized.

The embodiment of the invention provides a parallel operation method and device for a heterogeneous multi-core processor, a medium and a program product, and the method comprises the steps: generating an arithmetic unit corresponding to a function module according to the algorithm logic of the function module, the arithmetic unit comprises a first arithmetic unit deployed on the first core and / or a second arithmetic unit deployed on the second core; the function module is any one of a plurality of function modules, and the plurality of function modules can realize a target task when being executed according to a first time sequence; determining a second time sequence used for scheduling the arithmetic units corresponding to the plurality of functional modules, the second time sequence comprising at least one parallel time sequence, and the parallel time sequence scheduling at least one second arithmetic unit and at least one first arithmetic unit in parallel; and scheduling the operation unit according to the second time sequence. According to the embodiment of the invention, the processing time delay of the target task can be reduced, the processing efficiency of the processor is improved, and the load distribution on each processor core is improved.

The application relates to a simulation and calculation method for the coverage efficiency of a satellite networking area, and the method comprises the following steps: S100, editing and initializing a simulation scene; S200, parallelly simulating the space state of each satellite in the satellite networking through a plurality of simulation CPU threads, wherein the number of the satellite networking is one or more; S300, writing the data obtained after the editing and initializing of the simulation scene in step S100 and the space state data of each satellite obtained in step S200 into a GPU through at least one calculation CPU thread; and S400, parallelly calculating the coverage efficiency index of each satellite networking through the GPU. Through the heterogeneous multi-processor hybrid architecture of the multi-CPU and multi-GPU, the simulation part of the algorithm logic complexity but non-data intensive calculation is parallelly processed by the multi-CPU thread, and the simulation part of the algorithm logic simplicity but data intensive calculation is parallelly processed by the GPU, so that the calculation time is effectively reduced and the efficiency is effectively improved while the calculation precision is ensured.

The invention provides an intelligent medical object transfer trolley special for an operating room, and relates to the technical field of medical instrument data processing. The transfer trolley comprises a trolley body, a clean article bearing area, a waste storage area, a weighing sensing module and a central control unit, wherein the clean article bearing area and the waste storage area are arranged on the trolley body; the central control unit is configured to execute a surgical material closed loop verification logic. According to the novel intelligent transfer trolley, fluid interference can be overcome through algorithm logic, high-precision recognition is achieved through low-cost hardware, and safety verification logic based on physical data is constructed.

Examples disclosed herein may estimate locations of players not visible in a sporting broadcast video. A prediction model may be generated based on a training data set of in-venue tracking data that includes locations of all players at all times and the corresponding broadcast tracking data that may not necessarily contain the locations of all players at all times. The prediction model may be based on an algorithmic logic (e.g., a spline regression) or machine learning model (e.g., k-nearest neighbor, deep neural network). The generated predicted model may be used to estimate the unknown locations of players in broadcast tracking based on the known locations.

The embodiment of the invention provides a method for managing equipment direct connection peripherals, in the method, target equipment can be firstly mounted on a non-secure virtual machine manager, and then the non-secure virtual machine manager initiates an equipment donation request to a secure virtual machine manager. Therefore, the secure virtual machine manager does not need to execute operations such as importing a target device driver and the like, the complexity of the algorithm logic required to be executed by the secure world when the target device is mounted on the secure virtual machine manager is reduced, and the security of the secure virtual machine manager is improved on the basis of the security of the trusted environment guarantee device straight-through peripheral provided by the trusted execution environment subsystem. And the performance requirement on the trusted execution environment subsystem is reduced.

The invention discloses a shear wave data rapid processing and drawing method and system. The method comprises the steps that a first arrival file obtained through a wave velocity test is obtained and analyzed; acquiring preset parameter data information input by a user based on the parameterized input window; obtaining engineering information input by a user based on the engineering information input window; acquiring stratum layering condition information input by a user based on the stratum condition input window; in response to a calculation processing request of a user, performing automatic calculation according to a predefined algorithm logic to obtain a calculation result file including the wave velocities of compression waves and shear waves at each depth, the average wave velocity of the compression waves and the shear waves of each stratum, the equivalent shear wave velocity and the predominant period; and in response to an achievement map generation request of a user, automatically drawing and generating a wave velocity test achievement map according to the calculation result file and a preset format. According to the method, rapid calculation and automatic drawing are realized, the result accuracy is ensured, meanwhile, the time is saved, and the efficiency is improved.

Disclosed is a cryptographic execution-time governance infrastructure for artificial intelligence and digital systems, comprising a compute plane and a cryptographically isolated authority plane. Data is inseparably bound, at or before ingestion, to an execution authorization scope object and to an Algorithmic Logic Fingerprint (ALF). Computation may proceed in the compute plane, but irreversible effectuation, including output release, routing, settlement, actuation, or persistent state change, is withheld at an execution-finality boundary below the application layer unless the authority plane verifies that a runtime ALF matches the bound or approved ALF and satisfies an execution predicate. A cryptographic execution capability is released only upon successful verification. Ledger-Anchored Validation Receipts (LAVRs) are generated for both allow and deny outcomes prior to effectuation. The fail-closed architecture provides hardware-rooted policy enforcement and renders unauthorized digital execution technically impossible without requiring source-code disclosure or post-hoc auditing.

This invention discloses a Pirani data processing method based on singular value decomposition (SVD), belonging to the field of Pirani data processing technology. The method includes placing a Pirani vacuum gauge in an experimental environment for data acquisition and processing. The acquired data is integrated into an m×n dimensional matrix A. Singular value decomposition is performed, and principal components are extracted. Principal components are selected, and matrix truncation and reconstruction are performed to achieve dimensionality reduction reconstruction with a fixed k value. Post-processing is then performed to obtain calibration curves, and pressure interpolation calculations are performed. Multi-temperature point calibration and storage of calibration tables are also included. This invention utilizes a Pirani data processing method based on singular value decomposition. The largest singular value and its corresponding basic waveform, and the second largest singular value and its corresponding waveform, capture the most important correction pattern. The third and subsequent singular values ​​typically correspond to data fluctuations caused by random interference, measurement noise, etc. By forcibly setting them to zero, the algorithm logic becomes simple and stable.

The invention relates to an analysis method for a nuclear power equipment health model based on a multi-modal large model and electronic equipment, and the method comprises the steps: collecting multi-source data in an equipment management platform, and carrying out the training based on the multi-modal large model; wherein the multi-source data comprises work order feedback data and equipment operation data; obtaining a visual logic diagram of a to-be-detected model and corresponding node description information, performing cross-modal analysis by using the trained multi-modal large model, identifying and extracting node data in the model, restoring a complete algorithm logic of the model, and generating a logic topological graph; in combination with work order feedback data and equipment operation data, an operation process of a model is reproduced based on a logic topological graph and the equipment operation data, and attribution analysis is performed on alarm abnormity of the model by comparing actual output and ideal output of the model; and based on a result of the attribution analysis, generating an optimization suggestion of the model, and feeding back the optimization suggestion to a corresponding platform through an interface.

The application discloses a kind of deep neural network model application development systems based on deep learning, it is related to artificial intelligence technical field, the present application aims at solving the problems of gradient distortion caused by sample shunt in training and algorithm logic and bottom hardware load disconnection.System includes: saliency probe module, for generating sample saliency score;Bottom hardware direct connection controller, real-time monitoring hardware state and triggering computing power rescheduling instruction;Dynamic calculation path engine, according to the score, sample is distributed to full or sparse calculation path;Gradient alignment compensation module, scale compensation is carried out to sparse path gradient.Through the above-mentioned module cooperation, the present application constructs the real-time feedback link between algorithm and hardware, eliminates gradient magnitude fluctuation, improves hardware throughput while ensuring the stability and precision of model training.

The invention provides a multi-hole machining programming-free path planning method suitable for machining software, which is characterized in that a shortest distance tool path planning strategy in mainstream strategies is optimized, before path planning is carried out, all hole sites are subjected to region division, a plurality of secondary machining regions are divided according to information of to-be-machined holes in a total machining region range, and the secondary machining regions are subjected to path planning according to the information of the to-be-machined holes in the total machining region range; when the hole machining path planning is implemented, the next to-be-machined hole closest to the current hole can be found only by performing coordinate comparison, calculation and judgment on the to-be-machined hole in the adjacent secondary machining area and the current hole; by adopting the algorithm logic to calculate and judge the path planning of multi-hole processing, the number of hole sites to be processed by the system can be reduced, so that the rapid and automatic planning of the processing path is realized, compared with a traditional planning method, the method has the advantages that the efficiency of processing the calculated amount is effectively and greatly improved, the waiting time consumed by calculation is remarkably shortened, and the working efficiency is improved. The processing efficiency is improved, and the user experience is improved.