知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

105 records (took 0.028 seconds)

    METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS
    6.
    发明申请
    METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS 有权

    公开(公告)号：US20230032355A1

    公开(公告)日：2023-02-02

    申请号：US17963397

    申请日：2022-10-11

    Applicant: STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY

    Inventor： Jeffrey S. Myers ,  Kenneth J. Sanchez ,  Michael L. Bernico

    IPC: G06N3/04 ,  G06N3/08

    Abstract: A method of training and using a machine learning model that controls for consideration of undesired factors which might otherwise be considered by the trained model during its subsequent analysis of new data. For example, the model may be a neural network trained on a set of training images to evaluate an insurance applicant based upon an image or audio data of the insurance applicant as part of an underwriting process to determine an appropriate life or health insurance premium. The model is trained to probabilistically correlate an aspect of the applicant's appearance with a personal and/or health-related characteristic. Any undesired factors, such as age, sex, ethnicity, and/or race, are identified for exclusion. The trained model receives the image (e.g., a “selfie”) of the insurance applicant, analyzes the image without considering the identified undesired factors, and suggests the appropriate insurance premium based only on the remaining desired factors.

    METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS
    7.
    发明申请
    METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS 有权

    公开(公告)号：US20220156844A1

    公开(公告)日：2022-05-19

    申请号：US17591633

    申请日：2022-02-03

    Applicant: STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY

    Inventor： Jeffrey S. Myers ,  Kenneth J. Sanchez ,  Michael L. Bernico

    IPC: G06Q40/08 ,  G06V10/75 ,  G06V30/194 ,  G06V10/44 ,  G06N3/08 ,  H04N7/18

    Abstract: A method of training and using a machine learning model that controls for consideration of undesired factors which might otherwise be considered by the trained model during its subsequent analyses of new data. For example, the model may be a neural network trained on a set of training images to evaluate an insurance applicant based upon an image or audio data of the insurance applicant as part of an underwriting process to determine an appropriate life or health insurance premium. The model is trained to probabilistically correlate an aspect of the applicant's appearance with a personal and/or health-related characteristic. Any undesired factors, such as age, sex, ethnicity, and/or race, are identified for exclusion. The trained model receives the image (e.g., a “selfie”) of the insurance applicant, analyzes the image without considering the identified undesired factors, and suggests the appropriate insurance premium based only on the remaining desired factors.

    TECHNOLOGY FOR IMPLEMENTING A REVERSE COMMUNICATION SESSION FOR AUTOMOTIVE DEVICES
    8.
    发明申请
    TECHNOLOGY FOR IMPLEMENTING A REVERSE COMMUNICATION SESSION FOR AUTOMOTIVE DEVICES 有权

    公开(公告)号：US20220068043A1

    公开(公告)日：2022-03-03

    申请号：US17476356

    申请日：2021-09-15

    Applicant: STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY

    Inventor： Kenneth J. Sanchez ,  Eric Dahl ,  Steven J. Harris

    IPC: G07C5/00 ,  G07C5/08 ,  G08B27/00 ,  G08B25/10 ,  G06Q10/08

    Abstract: Systems and methods relate to, inter alia, receiving a set of sensor data from a set of sensors communicatively coupled to the electronic mobile device. The systems and methods may further include determining automatically from the set of sensor data that the collision event occurred during the operation of the automotive device. The systems and methods may further include initiating, in response to determining that the collision event occurred, a reverse communication session between the electronic mobile device and a collision management server. The electronic mobile device, via the initiated reverse communication session, may be configured to receive a first offer from a first service provider and a second offer from a second service provider. The first offer and the second offer are for servicing the automotive device by the first service provider and the second service provider, respectively.

Patent Agency Ranking