知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

11599 条记录 (耗时 0.043 秒)

    TRANSFER LEARNING TECHNIQUES FOR USING PREDICTIVE DIAGNOSIS MACHINE LEARNING MODELS TO GENERATE CONSULTATION RECOMMENDATION SCORES
    73.
    发明公开
    TRANSFER LEARNING TECHNIQUES FOR USING PREDICTIVE DIAGNOSIS MACHINE LEARNING MODELS TO GENERATE CONSULTATION RECOMMENDATION SCORES 审中-公开

    公开(公告)号：US20230153663A1

    公开(公告)日：2023-05-18

    申请号：US17530241

    申请日：2021-11-18

    申请人： UnitedHealth Group Incorporated

    发明人： Rick A. HAMILTON ,  Garry CHOY ,  Rafael Campos Do Amaral E VASCONCELLOS ,  Gregory J. BOSS

    IPC分类号： G06N7/00

    CPC分类号： G06N7/005

    摘要： Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing predictive data analysis operations. For example, certain embodiments of the present invention utilize systems, methods, and computer program products that perform predictive data analysis operations by an end-to-end machine learning framework that performs at least the following steps/operations: (i) a service request data object is processed by a diagnosis prediction machine learning model to generate a probabilistic diagnosis data object, (ii) the probabilistic diagnosis data object is processed by the hybrid diagnosis-provider classification machine learning model to generate a variable-length classification for the service request data object, and (iii) the variable-length classification is processed by a recommendation scoring machine learning model to generate a consultation recommendation score for the service request data object.

    METHOD AND SYSTEM FOR EVALUATION OF SYSTEM FAULTS AND FAILURES OF A GREEN ENERGY WELL SYSTEM USING PHYSICS AND MACHINE LEARNING MODELS
    74.
    发明公开
    METHOD AND SYSTEM FOR EVALUATION OF SYSTEM FAULTS AND FAILURES OF A GREEN ENERGY WELL SYSTEM USING PHYSICS AND MACHINE LEARNING MODELS 审中-公开

    公开(公告)号：US20230153661A1

    公开(公告)日：2023-05-18

    申请号：US17454663

    申请日：2021-11-12

    申请人： Banpu Innovation & Ventures LLC

    发明人： Othman Elkhomri

    IPC分类号： G06N7/00 ,  E21B47/07 ,  E21B47/10

    CPC分类号： G06N7/005 ,  E21B47/07 ,  E21B47/10 ,  E21B2200/20 ,  E21B2200/22

    摘要： A method of managing a well system includes: obtaining, by a digital twin manager and based on field well dynamics behavior data of the well system, emulated well dynamics behavior data using a physics-based model; obtaining, based on a predetermined monitoring criterion, predicted well dynamics behavior data of the well system using a physics constrained machine learning model that is based on the emulated well dynamics behavior data and the field well dynamics behavior data; determining, using a second machine learning model, an impact level that associates the predicted well dynamics behavior data with a well system abnormality; determining, using the second machine learning model, a likelihood level that associates the predicted well dynamics behavior data with the well system abnormality; determining a probability and a risk level of the well system abnormality based on the impact level and the likelihood level.

    GREEDY INFERENCE FOR RESOURCE-EFFICIENT MATCHING OF ENTITIES
    75.
    发明公开
    GREEDY INFERENCE FOR RESOURCE-EFFICIENT MATCHING OF ENTITIES 审中-公开

    公开(公告)号：US20230153382A1

    公开(公告)日：2023-05-18

    申请号：US17455046

    申请日：2021-11-16

    申请人： SAP SE

    发明人： Sundeep Gullapudi

    IPC分类号： G06K9/62 ,  G06N7/00

    CPC分类号： G06K9/6256 ,  G06N7/005 ,  G06K9/6262

    摘要： Methods, systems, and computer-readable storage media for determining a set of potential probability thresholds based on a set of inference results provided by processing testing data through the ML model, for each potential probability threshold in the set of potential probability thresholds, determining an accuracy, selecting a probability threshold from the set of potential probability thresholds, processing an inference job including sets of entity pairs through the ML model to assign a label to each entity pair in the sets of entity pairs, each label being associated with a probability and including a type of multiple types, and for each entity pair having a label of one or more specified types, selectively removing an entity of the entity pair from further processing of the inference job by the ML model based on whether the probability associated with the label meets or exceeds the probability threshold.

    Real-time predictions based on machine learning models
    76.
    发明授权
    Real-time predictions based on machine learning models 有权

    公开(公告)号：US11651291B2

    公开(公告)日：2023-05-16

    申请号：US16777686

    申请日：2020-01-30

    申请人： salesforce.com, inc.

    发明人： Rakesh Ganapathi Karanth ,  Arun Kumar Jagota ,  Kaushal Bansal ,  Amrita Dasgupta

    IPC分类号： G06N20/20 ,  G06N7/00

    CPC分类号： G06N20/20 ,  G06N7/005

    摘要： An online system performs predictions for real-time tasks and near real-time tasks that need to be performed by a deadline. A client device receives a real-time machine learning based model associated with a measure of accuracy. If the client device determines that a task can be performed using predictions having less than the specified measure of accuracy, the client device uses the real-time machine learning based model. If the client device determines that a higher level of accuracy of results is required, the client device sends a request to an online system. The online system provides a prediction along with a string representing a rationale for the prediction.

    Hardware-based machine learning acceleration
    77.
    发明授权
    Hardware-based machine learning acceleration 有权

    公开(公告)号：US11651260B2

    公开(公告)日：2023-05-16

    申请号：US16481816

    申请日：2018-01-31

    申请人： The Regents of the University of California

    发明人： Bita Darvish Rouhani ,  Mohammad Ghasemzadeh ,  Farinaz Koushanfar

    IPC分类号： G06N20/00 ,  G06N7/00 ,  G06F17/16 ,  G06N20/10

    CPC分类号： G06N7/005 ,  G06F17/16 ,  G06N20/00 ,  G06N20/10

    摘要： A method for hardware-based machine learning acceleration is provided. The method may include partitioning, into a first batch of data and a second batch of data, an input data received at a hardware accelerator implementing a machine learning model. The input data may be a continuous stream of data samples. The input data may be partitioned based at least on a resource constraint of the hardware accelerator. An update of a probability density function associated with the machine learning model may be performed in real time. The probability density function may be updated by at least processing, by the hardware accelerator, the first batch of data before the second batch of data. An output may be generated based at least on the updated probability density function. The output may include a probability of encountering a data value. Related systems and articles of manufacture, including computer program products, are also provided.

    System having multiple processing unit sets for training neural networks
    79.
    发明授权
    System having multiple processing unit sets for training neural networks 有权

    公开(公告)号：US11651226B2

    公开(公告)日：2023-05-16

    申请号：US16788918

    申请日：2020-02-12

    申请人： Graphcore Limited

    发明人： Helen Byrne ,  Luke Benjamin Hudlass-Galley ,  Carlo Luschi

    IPC分类号： G06F30/13 ,  G06N3/08 ,  G06K9/62 ,  G06N3/02 ,  G06N3/06 ,  G06N7/00

    CPC分类号： G06N3/08 ,  G06K9/6215 ,  G06K9/6256 ,  G06N3/02 ,  G06N3/06 ,  G06N7/00

    摘要： A data processing system for training a neural network, the data processing system comprising: a first set of one or more processing units running one model of the neural network, a second set of one or more processing units running another model of the neural network, a data storage, and an interconnect between the first set of one or more processing units, the second set of processing units and the data storage, wherein the data storage is configured to provide over the interconnect, training data to the first set of one or more processing units and the second set of one more processing units, wherein each of the first and second set of processing units is configured to, when performing the training, evaluate loss for the respective training iteration including a measure of the dissimilarity between the output values calculated based on the different modes running on the first and second set of processing units, wherein the dissimilarity measure is weighted in the evaluation of the loss in accordance with a parameter that is updated between different training iterations.

    MEMORY CONTROLLER AND MEMORY SYSTEM INCLUDING THE SAME
    80.
    发明公开
    MEMORY CONTROLLER AND MEMORY SYSTEM INCLUDING THE SAME 审中-公开

    公开(公告)号：US20230143905A1

    公开(公告)日：2023-05-11

    申请号：US17829669

    申请日：2022-06-01

    申请人： SAMSUNG ELECTRONICS CO., LTD.

    发明人： HOYOUN KIM

    IPC分类号： G11C11/406 ,  G06F3/06 ,  G06N7/00

    CPC分类号： G11C11/406 ,  G06F3/0655 ,  G06F3/0604 ,  G06F3/0679 ,  G06N7/005

    摘要： A memory controller, to control a semiconductor memory device, includes an access pattern profiler, a row hammer prediction neural network, and a memory interface. The access pattern profiler generates an access pattern profile based on a row access pattern on a portion of memory cell rows of the semiconductor memory device during a reference time interval posterior to a refresh interval during which the memory cell rows are refreshed. The row hammer prediction neural network predicts a probability of occurrence based on the access pattern profile. In response to the probability being equal to or greater than a reference value, the row hammer prediction neural network generates a hammer address, an alert signal indicating that the row hammer occurs, and an outcast row list. The memory interface transmits the hammer address, the outcast row list, and the alert signal to the semiconductor memory device.