268 results about "Local machine" patented technology

The invention discloses a method for generating and constructing a database test knowledge base based on an open source large language model and retrieval enhancement, and relates to the technical field of database testing, and the method comprises the steps: building a local deployment system on a local computer system or server, and achieving the efficient local operation of the open source large language model and the stable output of language processing service; on a local computer system or a server, a retrieval enhancement generation framework is constructed, and accurate knowledge support and reliable content output are provided for local language processing service by integrating multi-modal knowledge, optimizing retrieval and generation processes and guaranteeing safety and compliance; on a local computer system or a server, a personalized database knowledge base is built, database related information is collected and processed from multiple channels and is output in an adaptive form, and the specified information requirement of local personalized language processing service is met. According to the method, efficient and stable operation of the open source large language model is realized, accurate and reliable language processing output is realized, and personalized services meet specific requirements.

Techniques of updating machine learning models in a network include combining global and local models at each electronic entity of a network. For example, a first electronic entity (e.g., a user device) may train a local machine learning (ML) model based on data collected by the first electronic entity or other entities (e.g., other user devices, servers) and make predictions based on that model. Nevertheless, other electronic entities may also train or store their own local ML models. Accordingly, for more insight about the network and better predictability of the ML models, the first electronic entity may obtain a ML model from a second electronic entity, i.e., a global ML model. Upon receipt of the global ML model, the first electronic entity may aggregate the local ML model and the global ML model to produce an updated global ML model.

The systems and methods for improving pose estimation models are disclosed herein. Digital image data of an environment can be obtained and provided to a first machine learning model. A first confidence metric can be computed for the image. The first confidence metric can be compared with a threshold value and provided to a second machine learning model. A second confidence metric can be generated for training of machine learning models for pose estimation. A generic machine learning model can be updated using model parameters from trained local machine learning models.

An example method of displaying windows in a remote desktop system, includes: obtaining, by a remote desktop client executing on a local computer having a local operating system (OS), information relating a first window to a first virtual desktop generated by the local OS; sending the information from the remote desktop client to a remote desktop server; setting, by the remote desktop server, a tag in a remote window object representing a remote window that corresponds to the first window based on the information, the remote window generated by a remote OS on a remote computer; receiving, at the remote desktop client, at least a portion of the remote window object including the tag; and displaying, by the remote desktop client in cooperation with the local OS based on the tag, the first window on the first virtual desktop.

Systems and methods for implementing a balanced client selection for decentralized machine learning (BCS-DL) that can be utilized in both decentralized machine learning and hybrid machine learning for vehicle environments are described. For example, a vehicle can include a processor device training a machine learning model using local data, and a controller device performing balanced client selection in real-time to communicate the local machine learning model with a connected vehicle for decentralized machine learning. Hybrid machine learning combines aspects from federated learning and decentralized machine learning approaches. The disclosed BCS-DL system is designed to execute balanced client selection in real-time for vehicles that are acting as clients in a hybrid machine learning infrastructure. The balanced client selection also calculates a training contribution estimation (TCE) and a model weight computation (MWC) to mitigate imbalance in the data distribution incurred by non-Independent, Identically Distributed (non-IID data) related to decentralized and hybrid machine learning.

The invention discloses a multi-robot path coordination system based on reinforcement learning, and the system comprises a state collection module which is used for collecting the operation state and environment perception information of each robot, and generating an original data set; the state modeling module is used for constructing local state representation for each robot in each decision period; the opponent perception module is used for executing opponent learning perception training and outputting opponent parameter estimation vectors; the improved LOLA module is used for generating a final estimation result on the basis of self-strategy updating of the traditional LOLA; the strategy updating module is used for correcting the updating direction of the reinforcement learning strategy of the local machine and generating a feasible action set; the arbitration decision module is used for executing arbitration solution in combination with the candidate path action distribution to generate a scheduling result; and the execution feedback module is used for issuing the control action to a robot executor for execution, generating a path coordination result and storing the path coordination result. According to the invention, path coordination of multiple robots is realized.

In one embodiment, a supervisory device receives one or more goal parameters for trainer clients in a federated learning system configured to train local machine learning models using local datasets. The device configures, based on the one or more goal parameters, the trainer clients to quantize their local machine learning models prior to sending them for aggregation into a global model. The device determines an amount of resource savings associated with the trainer clients quantizing their local machine learning models. The device provides an indication of the amount of resource savings for presentation to a user.

The invention provides an unmanned aerial vehicle group obstacle avoidance method and system based on least square and distributed arbitration, and belongs to the field of multi-unmanned aerial vehicle path planning, and the method comprises the steps: maintaining the historical observation data of a dynamic obstacle; a trajectory prediction model is constructed based on historical observation data, and a future trajectory prediction set is generated through least square fitting; selecting an optimal prediction model from the trajectory prediction set through a backtracking verification strategy; on the basis of a preset distributed arbitration rule, whether the local machine obtains the release right of the obstacle prediction information or not is judged; packaging the trajectory information corresponding to the optimal prediction model into a standard message by using the unmanned aerial vehicle which obtains the release right, and broadcasting the standard message to the cluster; and incorporating the obtained predicted trajectory message into collision avoidance constraint of a planner, performing local trajectory planning, and generating a collision-free flight path. According to the method, the safety, the collaboration and the passing efficiency of the cluster system in a real complex scene are remarkably improved.

The invention discloses a historical record backtracking method based on LLM, and relates to the field of data retrieval and natural language processing, and the method comprises the steps: collecting various operation data of a user in a local computer, extracting text information in pictures and voices in combination with an OCR technology and an ASR technology, and storing the text information and related information in a database in a unified format; the method comprises the following steps of: processing collected multi-modal data information, firstly performing de-duplication and cleaning on the collected data, then converting a collected screenshot into a video, and finally performing vector embedding on a video image frame, an OCR (Optical Character Recognition) result and a voice transcription text and storing the video image frame, the OCR result and the voice transcription text into a vector database; optimizing the pre-trained LLM in a specific direction by adopting a fine tuning technology; according to the method, conversation with a historical timeline is realized, user input is captured and stored in a query queue, personalized response is generated through LLM interaction in combination with a memory module and an RAG technology, and answers and related image or audio links are displayed.

Embodiments are disclosed for a group security scheme for an autonomous vehicle engaged in a collaborative machine learning approach. As an example, a method comprises: generating, in a vehicle, a digital signature based on a first key and a second key, both of the first key and the second key received from a group manager, and transmitting a message signed with the digital signature to a collaborator, the message including coefficients of a local machine learning model of the vehicle. In this way, an accuracy of the local machine learning model of the vehicle may be increased, while a privacy of the vehicle is increased.

A local machine learning based recording and audio transcribing system is proposed that operates in real-time in parallel and time-synchronized with a operating theatre recorder system, modifying encoding of the operating theatre recorder system outputs for generation of the pre-processed recording files to be transmitted to a remote cloud-based processing backend using operating a centralized machine learning model data architecture across a network. By modifying the recording or the generation of the pre-processed recording files, the system can be tuned for providing increased signal resolution at more relevant portions of time or frame portions to improve the accuracy and predictive capability of a backend machine learning model that is configured for operation using the pre-processed recording files while adapting for practical network bandwidth and computing resource limitations in cloud-based implementations.

The invention relates to a kernel-based federal learning method and system, computer equipment and a storage medium. Target clients are selected from a client set to form a client subset; sending a first parameter of the first machine learning model and the number of iterations to the client subset; wherein the target client is used for performing random feature sampling and double stochastic gradient evaluation according to the first parameter and the number of iterations, and updating a second machine learning model based on double stochastic gradients to generate a second parameter; the first machine learning model is a local machine learning model of the server, and the second machine learning model is a local machine learning model of the target client; receiving a second parameter returned by the client subset, and updating the first machine learning model according to the second parameter; the expansibility of the kernel method in a federated learning scene is improved; and model convergence acceleration is facilitated.

The invention provides a security and protection monitoring system based on a master-slave controller mode. The system comprises a master controller elected in advance through an election mechanism, and a plurality of slave controllers in communication connection with the master controller, each slave controller receives and processes a video processing task sent by the video monitoring equipment bound with the slave controller, and calculates a current computing power load according to the number of video frames which can be processed and output by a local machine per second, the number of targets needing to be processed in each frame of image and the number of tasks waiting to be processed in a task queue; according to the technical scheme of the invention, the current computing power load is sent to the master controller, so that the master controller can determine the overload slave controller and the low-load slave controller according to the current computing power load sent by each slave controller, and distributes a part of video processing tasks in the overload slave controller to the low-load slave controller, thereby reducing the response delay and improving the video processing efficiency. And system crash can be avoided, and computing power cooperation among the slave controllers is realized, so that the overall performance of the system is effectively improved.

The invention discloses a multi-cluster hot debugging method based on Service Mesh, and belongs to the technical field of cloud native software development. Through a debugging instruction, a multi-cluster intelligent routing gateway establishes a unified and logically isolated debugging session and a virtual network environment for a developer terminal and a plurality of target Kubernetes clusters; the gateway dynamically allocates a local port for a debugging task from a global resource pool, and instructs a target cluster to create a Pod; the gateway serves as a center node, performs cross-cluster routing conflict detection, generates a globally consistent dynamic routing rule and issues the dynamic routing rule to a target cluster; and finally, forwarding the online traffic matched with the rule to an appointed port of a developer local machine through a proxy Pod and a gateway. Through a centralized intelligent coordination architecture, various conflicts in a multi-cluster environment are fundamentally avoided, and the debugging efficiency, stability and automation level are remarkably improved.

Systems and methods for federated model validation and data verification are disclosed. A method may include: (1) receiving, by a local computer program executed by client system, a federated machine learning model from a federated model server; (2) testing, by the local computer program and using a policy service, the federated machine learning model for vulnerabilities to attacks; (3) accepting, by the local computer program, the federated machine learning model in response to the federated machine learning model passing the testing; (4) training, by the local computer program, the federated machine learning model using input data comprising local data and outputting training parameters; (5) identifying, by the local computer program using the policy service, accidental leakage and / or contamination by comparing the training parameters to the input data; and (6) providing, by the local computer program, the training parameters to the federated model server.

This disclosure provides a method for machine learning model training in a communications network. The method comprises transmitting from a first network node to a second network node a request for a first indication of a label, a second indication of intermediate training results, and / or a third indication of a loss calculation; and receiving at the first network node from the second network node a first indication of the label, a second indication of the intermediate training results, and / or a third indication of the calculated loss; determining at the first network node a model convergence based on the received first indication, the second indication, and / or the third indication, and initiating at the first network node an update of a local Machine Learning (ML) model using the at least one intermediate training result and / or the third indication of the calculated loss; particularly wherein the initiating comprises storing at the first network node the ML model.

The invention provides a master-slave-free control method and device of a liquid cooling CDU parallel operation system, a CDU, a system and a medium, and belongs to the field of liquid cooling heat dissipation. The parallel operation system comprises at least two liquid cooling CDUs, and each liquid cooling CDU is provided with a liquid uniformizing control circuit. The liquid homogenizing control circuit detects a deviation signal between a local flow signal and an average flow signal; the method is applied to any liquid cooling CDU, and comprises the following steps: acquiring a deviation signal of a local machine based on a liquid equalizing control circuit; based on the deviation signal of the local machine, determining a uniform liquid compensation signal; acquiring a flow reference signal, and generating a machine control signal based on the flow reference signal and the liquid equalizing compensation signal; and performing liquid homogenizing control on the machine according to the control signal of the machine. Even if any liquid cooling CDU breaks down or communication between the liquid cooling CDUs goes wrong, work of other liquid cooling CDUs is not affected, the cooling effect of the liquid cooling CDU parallel operation system can still be kept, and the phenomenon that equipment needing to be cooled is overheated is avoided.

A computing device, that is configured to configure a global machine learning model, performs respective electronic risk audits of client devices configured to train respective local machine learning models that correspond to a global machine learning model. Based on respective electronic risk scores of one or more of the client devices, determined via the respective electronic risk audits, the computing device implements one or more parameter privacy adjustment methods on respective parameters received from the client devices prior to using the respective parameters to configure the global machine learning model, wherein respective client devices determined to have higher electronic risk scores have more of the parameter privacy adjustment methods applied than other respective client devices determined to have lower electronic risk scores. The computing device provides, to the client devices, the global machine learning model configured according to the respective parameters as adjusted.

A system for data capture and storage comprising: a data capture device taken from the group consisting of blood pressure device, thermometer, pulse sensor, scale and glucose meter; a local computer system in communication with the data capture device; a set of local computer readable instructions configured for receiving data from the data capture device, display the data capture device in a user-friendly format, determine if there is an electronic communications link and if so, transmit the data to a remote computer system, retrieve data from the remote computer system and display a combination of the data frohe data capture device and the remote computer system to the user.

Methods and a system for a data pool which allows for local processing of database data and a server which is ignorant of data content and encryption keys for the database data. The data pool provides for local decryption and processing of data pool data, reducing security vulnerabilities and server load. The data pool also includes methods for creating an audit trail on a remote server of changes made to the data pool data locally so that in the event of an unexpected disconnection or shutdown of the local machine without uploading updated data, changes to the data pool data may be recreated on a subsequent use session with the local machine with minimal loss to changes or updates made in a previous session.

The invention concerns a method that includes a computation loop including a transmission step for computing, by a central node, an output of a current aggregated model for each image of a central dataset. The method also includes transferring, to each of n local nodes, data representative of the aggregated model; and a set of prototypes. The method also includes a training step including for each local node, training a respective local computer vision model to obtain a respective trained local model; and performing supervised training of the aggregated model, based on the central dataset, thereby obtaining a trained central model. The method also includes an aggregation step including for each local node, transferring, to the central node, corresponding local model data; and updating the aggregated model based on the local model data; and data representative of the trained central model.

Disclosed are systems and methods for enabling a developer to use a local browser, running on a local machine of the developer in a first location to access real devices (e.g., smart phones) at a second location, such as a data center. The developer can select and control the remote devices, in the second location. The described embodiments can capture developer's inputs from the first location and input them to the remote device in the second location. A video stream of the remote device is transmitted to the browser in the first location and displayed in a replica canvas on the developer's browser. The developer can interact with the canvas on his / her browser, as if the remote device were present in the first location.

This invention discloses a data processing method, system, device, and medium based on a user-selected local model. The method includes: acquiring a local machine learning model file in response to a user operation; dynamically parsing model metadata to determine input and output specifications; acquiring multimodal target data and preprocessing and adapting it according to the input specifications; calling local hardware resources to execute model inference; parsing the inference results according to the output specifications and outputting them to the target business system. This invention overcomes the closed nature of traditional software-built-in AI models, achieving full utilization of local AI computing power and effective privacy protection of sensitive data. Through a highly abstract underlying adaptation architecture, this invention is not dependent on specific application scenarios and can be widely integrated as a general-purpose underlying technology into various industry software such as video restoration, medical image analysis, professional image retouching, and industrial inspection, giving the software flexible AI capability scalability.

The invention discloses a crystal oscillator-free Bluetooth chip pairing method and device and a communication system, and the method comprises the steps: switching a preset frequency offset value for polling between a paging scanning state and a query scanning state, so as to search an ID packet sent by host equipment in the query state; when the ID packet is received, a slave equipment information packet of the host is sent to the host equipment; receiving a host equipment information packet sent by the host equipment in response to the slave equipment information packet, so that the host equipment can establish Bluetooth connection with a local machine; and sending a connection request to the host device according to the address information of the host device loaded in the host device information packet in the paging state to establish a Bluetooth connection with the host device so as to realize Bluetooth pairing between the local machine and the host device. For low-power-consumption equipment, the Bluetooth chip without the crystal oscillator is ensured to be paired with other equipment to establish connection. The power consumption of low-power-consumption equipment is reduced, the development difficulty, development cost and period of products are reduced, and the equipment is more miniaturized.

System that features the following: a processing unit that is functionally connected to a memory; an AI (Artificial Intelligence) platform that exchanges data with the processing unit, wherein the AI ​​platform is designed to train a machine learning (ML) model, wherein the AI ​​platform has the following features: a registration manager to register participating entities in a cooperative relationship, to arrange the registered entities in a topology, and to establish a topological data transmission protocol; an encryption manager to generate a public additive homomorphic encryption key, AHE key, and distribute it to each registered entity; an entity manager to control direct local encryption of machine learning model weights of local entities on registered entities using a distributed public AHE key, to selectively aggregate the encrypted local machine learning model weights including distributing the aggregated encrypted weights to one or more other participating entities in the topology, or to support local aggregation on the registered entities according to the topological data transmission protocol; where the encryption manager decrypts an aggregated sum of the encrypted local machine learning model weights with a corresponding private AHE key, and distributes the decrypted aggregated sum to each entity in the topology.

An integrated computer network security and threat prevention and detection platform includes a central processor and a display operable to aggregate and present data from a plurality of network security applications in an integrated dashboard format to a system administrator. The network security applications may be hardware, software, or hybrid applications running on local machines, local networks, remote machines, or remote networks, in communication with the central processor. In one embodiment implementation of the integrated computer network security and threat prevention and detection platform is performed on premises, in an alternative embodiment the integrated computer network security and threat prevention and detection platform is provided in an Internet or cloud-based environment, in other embodiments the computer system security platform is a hybrid configuration having both on-premises and cloud base components.

Methods and a system for a data pool which allows for local processing of database data and a server which is ignorant of data content and encryption keys for the database data. The data pool provides for local decryption and processing of data pool data, reducing security vulnerabilities and server load. The data pool also includes methods for creating an audit trail on a remote server of changes made to the data pool data locally so that in the event of an unexpected disconnection or shutdown of the local machine without uploading updated data, changes to the data pool data may be recreated on a subsequent use session with the local machine with minimal loss to changes or updates made in a previous session.

Methods and systems for automating surgical procedures in model organisms. The system includes a user input panel, a trained computer vision system, a server, and articulated robotic arm. The system may further include an analytical scale, an anesthesia chamber, a tube labeling system, a tissue freezing system, and a biohazard waste disposal system. The computer vision system communicates with the robotic arm to coordinate tissue collections and common research procedures such as injections via recognition models related to detection, identification, localization, and classification. The central server provides rapid synchronization of data between the local machine and a cloud account system for real-time data analytics. The anesthesia chamber provides standardized administration of anesthesia, the sample labeling system provides 2D barcoded labels according to user input, and the tissue freezing and waste disposal systems enable storage of samples and removal of carcasses following tissue collection, fully automating the necropsy.

This invention discloses a federated averaging algorithm based on residual networks, characterized by high accuracy and efficient communication: Step 1: The server partitions the MNIST dataset into IID and Non-IID categories and distributes them to various clients; Step 2: Each client locally constructs a residual network model; Step 3: The number of communication rounds R between the server and clients is set. In each round, the server randomly selects K clients to participate in the communication and sends the parameters of the global model; Step 4: The selected clients download the parameters of the global model and perform local machine learning training to obtain the trained model parameters; Step 5: The K clients upload the trained model parameters to the server; Step 6: The server updates the global model parameters according to a weighted average aggregation strategy and repeats steps 3-6 until the final global model parameters are obtained in the Rth round of communication. This invention significantly improves accuracy and communication efficiency compared to existing technologies.

Techniques for training a global model for use at a host of oilfield application sites based on raw data obtained from the application sites without direct exposure of the data to the global model. The techniques include developing and distributing a global model with a predetermined set of parameter weights. The model is then locally employed at each application site by a local computer which maintains the integrity of the acquired data during performance of the oilfield application. The data is used to update the parameter weights based on real-time circumstances. Thus, the parameter weights may be transmitted to the centralized computer for updating of the global model. Further, the updated global model may continue to direct other applications and the process continued in a beneficial feedback loop manner.