中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

8 records (took 0.016 seconds)

    META-LEARNING FOR MULTI-TASK LEARNING FOR NEURAL NETWORKS
    1.
    发明申请
    META-LEARNING FOR MULTI-TASK LEARNING FOR NEURAL NETWORKS 有权

    公开(公告)号：US20210406609A1

    公开(公告)日：2021-12-30

    申请号：US17344758

    申请日：2021-06-10

    Applicant: Magic Leap, Inc.

    Inventor： Andrew Rabinovich ,  Vijay Badrinarayanan ,  Srivignesh Rajendran ,  Chen-Yu Lee

    IPC: G06K9/62 ,  G06N3/04 ,  G06N3/08

    Abstract: Methods and systems for meta-learning are described for automating learning of child tasks with a single neural network. The order in which tasks are learned by the neural network can affect performance of the network, and the meta-learning approach can use a task-level curriculum for multi-task training. The task-level curriculum can be learned by monitoring a trajectory of loss functions during training. The meta-learning approach can learn to adapt task loss balancing weights in the course of training to get improved performance on multiple tasks on real world datasets. Advantageously, learning to dynamically balance weights among different task losses can lead to superior performance over the use of static weights determined by expensive random searches or heuristics. Embodiments of the meta-learning approach can be used for computer vision tasks or natural language processing tasks, and the trained neural networks can be used by augmented or virtual reality devices.

    Meta-learning for multi-task learning for neural networks
    2.
    发明授权
    Meta-learning for multi-task learning for neural networks 有权

    公开(公告)号：US11048978B2

    公开(公告)日：2021-06-29

    申请号：US16185582

    申请日：2018-11-09

    Applicant: Magic Leap, Inc.

    Inventor： Andrew Rabinovich ,  Vijay Badrinarayanan ,  Srivignesh Rajendran ,  Chen-Yu Lee

    IPC: G06K9/62 ,  G06N3/04 ,  G06N3/08

    Abstract: Methods and systems for meta-learning are described for automating learning of child tasks with a single neural network. The order in which tasks are learned by the neural network can affect performance of the network, and the meta-learning approach can use a task-level curriculum for multi-task training. The task-level curriculum can be learned by monitoring a trajectory of loss functions during training. The meta-learning approach can learn to adapt task loss balancing weights in the course of training to get improved performance on multiple tasks on real world datasets. Advantageously, learning to dynamically balance weights among different task losses can lead to superior performance over the use of static weights determined by expensive random searches or heuristics. Embodiments of the meta-learning approach can be used for computer vision tasks or natural language processing tasks, and the trained neural networks can be used by augmented or virtual reality devices.

    Room layout estimation methods and techniques
    4.
    发明授权
    Room layout estimation methods and techniques 有权

    公开(公告)号：US10657376B2

    公开(公告)日：2020-05-19

    申请号：US15923511

    申请日：2018-03-16

    Applicant: Magic Leap, Inc.

    Inventor： Chen-Yu Lee ,  Vijay Badrinarayanan ,  Tomasz Malisiewicz ,  Andrew Rabinovich

    IPC: G06K9/00 ,  G06N3/04 ,  G06K9/46 ,  G06K9/34 ,  G06K9/62 ,  G06N3/08

    Abstract: Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.

    Meta-learning for multi-task learning for neural networks
    5.
    发明授权
    Meta-learning for multi-task learning for neural networks 有权

    公开(公告)号：US11853894B2

    公开(公告)日：2023-12-26

    申请号：US17344758

    申请日：2021-06-10

    Applicant: Magic Leap, Inc.

    Inventor： Andrew Rabinovich ,  Vijay Badrinarayanan ,  Srivignesh Rajendran ,  Chen-Yu Lee

    IPC: G06N3/084 ,  G06F18/21 ,  G06N3/04 ,  G06N3/044 ,  G06N3/047

    CPC classification number: G06N3/084 ,  G06F18/217 ,  G06N3/04 ,  G06N3/044 ,  G06N3/047

    Abstract: Methods and systems for meta-learning are described for automating learning of child tasks with a single neural network. The order in which tasks are learned by the neural network can affect performance of the network, and the meta-learning approach can use a task-level curriculum for multi-task training. The task-level curriculum can be learned by monitoring a trajectory of loss functions during training. The meta-learning approach can learn to adapt task loss balancing weights in the course of training to get improved performance on multiple tasks on real world datasets. Advantageously, learning to dynamically balance weights among different task losses can lead to superior performance over the use of static weights determined by expensive random searches or heuristics. Embodiments of the meta-learning approach can be used for computer vision tasks or natural language processing tasks, and the trained neural networks can be used by augmented or virtual reality devices.

    ROOM LAYOUT ESTIMATION METHODS AND TECHNIQUES
    6.
    发明申请
    ROOM LAYOUT ESTIMATION METHODS AND TECHNIQUES 审中-公开

    公开(公告)号：US20200234051A1

    公开(公告)日：2020-07-23

    申请号：US16844812

    申请日：2020-04-09

    Applicant: Magic Leap, Inc.

    Inventor： Chen-Yu Lee ,  Vijay Badrinarayanan ,  Tomasz Jan Malisiewicz ,  Andrew Rabinovich

    IPC: G06K9/00 ,  G06N3/04 ,  G06K9/46 ,  G06K9/34 ,  G06K9/62 ,  G06N3/08

    Abstract: Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.

    ROOM LAYOUT ESTIMATION METHODS AND TECHNIQUES
    8.
    发明申请
    ROOM LAYOUT ESTIMATION METHODS AND TECHNIQUES 审中-公开

    公开(公告)号：US20180268220A1

    公开(公告)日：2018-09-20

    申请号：US15923511

    申请日：2018-03-16

    Applicant: Magic Leap, Inc.

    Inventor： Chen-Yu Lee ,  Vijay Badrinarayanan ,  Tomasz Malisiewicz ,  Andrew Rabinovich

    IPC: G06K9/00 ,  G06N3/04 ,  G06K9/34 ,  G06K9/46

    Abstract: Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.

Patent Agency Ranking