中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

16 records (took 0.013 seconds)

    Unsupervised detection of intermediate reinforcement learning goals
    2.
    发明授权
    Unsupervised detection of intermediate reinforcement learning goals 有权

    公开(公告)号：US12106200B2

    公开(公告)日：2024-10-01

    申请号：US18168000

    申请日：2023-02-13

    Applicant: Google LLC

    Inventor： Pierre Sermanet

    IPC: G06N3/006 ,  G06F18/21 ,  G06F18/2111 ,  G06F18/2132 ,  G06N3/045 ,  G06N3/08

    CPC classification number: G06N3/006 ,  G06F18/2111 ,  G06F18/2132 ,  G06F18/217 ,  G06N3/045 ,  G06N3/08

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for detecting intermediate reinforcement learning goals. One of the methods includes obtaining a plurality of demonstration sequences, each of the demonstration sequences being a sequence of images of an environment while a respective instance of a reinforcement learning task is being performed; for each demonstration sequence, processing each image in the demonstration sequence through an image processing neural network to determine feature values for a respective set of features for the image; determining, from the demonstration sequences, a partitioning of the reinforcement learning task into a plurality of subtasks, wherein each image in each demonstration sequence is assigned to a respective subtask of the plurality of subtasks; and determining, from the feature values for the images in the demonstration sequences, a respective set of discriminative features for each of the plurality of subtasks.

    MIRROR LOSS NEURAL NETWORKS
    4.
    发明申请
    MIRROR LOSS NEURAL NETWORKS 有权

    公开(公告)号：US20230020615A1

    公开(公告)日：2023-01-19

    申请号：US17893454

    申请日：2022-08-23

    Applicant: Google LLC

    Inventor： Pierre Sermanet

    IPC: B25J9/16 ,  G05B13/02 ,  G06N3/08

    Abstract: This description relates to a neural network that has multiple network parameters and is configured to receive an input observation characterizing a state of an environment and to process the input observation to generate a numeric embedding of the state of the environment. The neural network can be used to control a robotic agent. The network can be trained using a method comprising: obtaining a first observation captured by a first modality; obtaining a second observation that is co-occurring with the first observation and that is captured by a second, different modality; obtaining a third observation captured by the first modality that is not co-occurring with the first observation; determining a gradient of a triplet loss that uses the first observation, the second observation, and the third observation; and updating current values of the network parameters using the gradient of the triplet loss.

    Unsupervised detection of intermediate reinforcement learning goals
    7.
    发明授权
    Unsupervised detection of intermediate reinforcement learning goals 有权

    公开(公告)号：US11580360B2

    公开(公告)日：2023-02-14

    申请号：US16347651

    申请日：2017-11-06

    Applicant: GOOGLE LLC

    Inventor： Pierre Sermanet

    IPC: G06N3/045 ,  G06N3/08 ,  G06F18/21 ,  G06F18/2111 ,  G06F18/2132

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for detecting intermediate reinforcement learning goals. One of the methods includes obtaining a plurality of demonstration sequences, each of the demonstration sequences being a sequence of images of an environment while a respective instance of a reinforcement learning task is being performed; for each demonstration sequence, processing each image in the demonstration sequence through an image processing neural network to determine feature values for a respective set of features for the image; determining, from the demonstration sequences, a partitioning of the reinforcement learning task into a plurality of subtasks, wherein each image in each demonstration sequence is assigned to a respective subtask of the plurality of subtasks; and determining, from the feature values for the images in the demonstration sequences, a respective set of discriminative features for each of the plurality of subtasks.

    Mirror loss neural networks
    9.
    发明授权
    Mirror loss neural networks 有权

    公开(公告)号：US11453121B2

    公开(公告)日：2022-09-27

    申请号：US16468987

    申请日：2018-03-19

    Applicant: Google LLC

    Inventor： Pierre Sermanet

    IPC: G05B19/04 ,  B25J9/16 ,  G05B13/02 ,  G06N3/08 ,  H04N7/18

    Abstract: This description relates to a neural network that has multiple network parameters and is configured to receive an input observation characterizing a state of an environment and to process the input observation to generate a numeric embedding of the state of the environment. The neural network can be used to control a robotic agent. The network can be trained using a method comprising: obtaining a first observation captured by a first modality; obtaining a second observation that is co-occurring with the first observation and that is captured by a second, different modality; obtaining a third observation captured by the first modality that is not co-occurring with the first observation; determining a gradient of a triplet loss that uses the first observation, the second observation, and the third observation; and updating current values of the network parameters using the gradient of the triplet loss.

    CONTROLLING AGENTS USING LATENT PLANS
    10.
    发明申请
    CONTROLLING AGENTS USING LATENT PLANS 有权

    公开(公告)号：US20220076099A1

    公开(公告)日：2022-03-10

    申请号：US17432366

    申请日：2020-02-19

    Applicant: Google LLC

    Inventor： Pierre Sermanet ,  Seyed Mohammad Khansari Zadeh ,  Harrison Corey Lynch

    IPC: G06N3/04 ,  G06N3/08

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling an agent. One of the methods includes controlling the agent using a policy neural network that processes a policy input that includes (i) a current observation, (ii) a goal observation, and (iii) a selected latent plan to generate a current action output that defines an action to be performed in response to the current observation.

Patent Agency Ranking