利索-全球专利检索

  1. Login
  2. Sign Up

Search Results

86 records (took 0.024 seconds)

    METHOD AND APPARATUS FOR GENERATING TRAINING DATA FOR GRAPH NEURAL NETWORK
    1.
    发明公开
    METHOD AND APPARATUS FOR GENERATING TRAINING DATA FOR GRAPH NEURAL NETWORK 审中-公开

    公开(公告)号：US20240078436A1

    公开(公告)日：2024-03-07

    申请号：US18259563

    申请日：2021-01-04

    Applicant: Robert Bosch GmbH ,  TSINGHUA UNIVERSITY

    Inventor： Hang Su ,  Jun Zhu ,  Zhengyi Wang ,  Hao Yang

    IPC: G06N3/094

    CPC classification number: G06N3/094

    Abstract: A method for generating adversarial examples for a Graph Neural Network (GNN) model. The method includes: determining vulnerable features of target nodes in a graph based on querying the GNN model, wherein the graph comprising nodes including the target nodes and edges, each of the edges connecting two of the nodes; grouping the target nodes into a plurality of clusters according to the vulnerable features of the target nodes; and obtaining the adversarial examples based on the plurality of clusters.

    Method and Apparatus for Weight-Sharing Neural Network with Stochastic Architectures
    2.
    发明公开
    Method and Apparatus for Weight-Sharing Neural Network with Stochastic Architectures 审中-公开

    公开(公告)号：US20240037390A1

    公开(公告)日：2024-02-01

    申请号：US18249162

    申请日：2020-10-15

    Applicant: Robert Bosch GmbH ,  Tsinghua University

    Inventor： Jun Zhu ,  Zhijie Deng ,  Yinpeng Dong ,  Chao Zhang ,  Kevin Yang

    IPC: G06N3/08 ,  G06N3/047

    CPC classification number: G06N3/08 ,  G06N3/047

    Abstract: A method for training a weight-sharing neural network with stochastic architectures is disclosed. The method includes (i) selecting a mini-batch from a plurality of mini-batches, a training data set for a task being grouped into the plurality of mini-batches and each of the plurality of mini-batches comprising a plurality of instances: (ii) stochastically selecting a plurality of network architectures of the neural network for the selected mini-batch; (iii) obtaining a loss for each instance of the selected mini-batch by applying the instance to one of the plurality of network architectures; and (iv) updating shared weights of the neural network based on the loss for each instance of the selected mini-batch.

    METHOD AND APPARATUS FOR DEEP LEARNING
    4.
    发明公开
    METHOD AND APPARATUS FOR DEEP LEARNING 审中-公开

    公开(公告)号：US20240256889A1

    公开(公告)日：2024-08-01

    申请号：US18565510

    申请日：2021-05-31

    Applicant: ROBERT BOSCH GMBH ,  TSINGHUA UNIVERSITY

    Inventor： Hang Su ,  Jun Zhu ,  Tianyu Pang ,  Xiao Yang ,  Yinpeng Dong ,  Zhijie Deng ,  Ze Cheng

    IPC: G06N3/094

    CPC classification number: G06N3/094

    Abstract: A method for deep learning. The method includes: receiving, by a deep learning model, a plurality of samples and a plurality of labels corresponding to the plurality of samples; adversarially augmenting, by the deep learning model, the plurality of samples based on a threat model; and assigning, by the deep learning model, a low predictive confidence to one or more adversarially augmented samples of the plurality of adversarially augmented samples having noisy labels due to the adversarially augmenting based on the threat model.

    METHOD AND APPARATUS FOR DEEP NEURAL NETWORKS HAVING ABILITY FOR ADVERSARIAL DETECTION
    5.
    发明公开
    METHOD AND APPARATUS FOR DEEP NEURAL NETWORKS HAVING ABILITY FOR ADVERSARIAL DETECTION 审中-公开

    公开(公告)号：US20240086716A1

    公开(公告)日：2024-03-14

    申请号：US18263576

    申请日：2021-02-26

    Applicant: Robert Bosch GmbH ,  TSINGHUA UNIVERSITY

    Inventor： Hang Su ,  Jun Zhu ,  Zhijie Deng ,  Ze Cheng

    IPC: G06N3/094

    CPC classification number: G06N3/094

    Abstract: A method for training a deep neural network (DNN) capable of adversarial detection. The DNN is configured with a plurality of sets of weights candidates. The method includes inputting training data selected from training data set to the DNN. The method further includes calculating, based on the training data, a first term for indicating a difference between a variational posterior probability distribution and a true posterior probability distribution of the DNN. The method further includes perturbing the training data to generate perturbed training data; and calculating a second term for indicating a quantification of predictive uncertainty on the perturbed training data. The method further includes updating the plurality of sets of weights candidates of the DNN based on augmenting the summation of the first term and the second term.

    METHOD AND APPARATUS FOR VISUAL REASONING
    6.
    发明公开
    METHOD AND APPARATUS FOR VISUAL REASONING 审中-公开

    公开(公告)号：US20240185023A1

    公开(公告)日：2024-06-06

    申请号：US18546842

    申请日：2021-03-03

    Applicant: Robert Bosch GmbH ,  TSINGHUA UNIVERSITY

    Inventor： Bo Zhang ,  Chongxuan Li ,  Hang Su ,  Jun Zhu ,  Ke Su ,  Siliang Lu ,  Ze Cheng

    IPC: G06N3/042 ,  G06N3/08

    CPC classification number: G06N3/042 ,  G06N3/08

    Abstract: A method for visual reasoning. The method includes: providing a network with sets of inputs and sets of outputs, wherein each set of inputs of the sets of inputs mapping to one of a set of outputs corresponding to the set of inputs based on visual information on the set of inputs, and wherein the network comprising a Probabilistic Generative Model (PGM) and a set of modules; determining a posterior distribution over combinations of one or more modules of the set of modules through the PGM, based on the provided sets of inputs and sets of outputs; and applying domain knowledge as one or more posterior regularization constraints on the determined posterior distribution.

    Illumination system with freeform surface
    8.
    发明授权
    Illumination system with freeform surface 有权

    公开(公告)号：US10527255B2

    公开(公告)日：2020-01-07

    申请号：US15787718

    申请日：2017-10-19

    Applicant: Tsinghua University ,  HON HAI PRECISION INDUSTRY CO., LTD.

    Inventor： Jun Zhu ,  Xiao-Fei Wu ,  Guo-Fan Jin ,  Shou-Shan Fan

    IPC: F21V5/04

    Abstract: An illumination system with freeform surface comprises a plurality of collimated light sources having same parameters and a freeform surface lens comprising a first freeform surface and a second freeform surface, wherein formula of the first freeform surface and the second freeform surface is expressed as follows: z = c ⁡ ( x 2 + y ⁢ ⁢ 2 ) 1 + 1 - ( 1 + k ) ⁢ c 2 ⁡ ( x 2 + y 2 ) + ∑ m ⁢ ∑ n ⁢ A mn ⁢ x m ⁢ y n , in which c is the curvature of the conic surface at the vertex, k is the conic constant, Amn represents the xy polynomials coefficient, m+n≥2 and both m and n are even, beams emitted by the plurality of collimated light sources pass through the freeform surface lens to form a plurality of light spots on a target plane.

    Oblique camera lens
    10.
    发明授权
    Oblique camera lens 有权

    公开(公告)号：US10386619B2

    公开(公告)日：2019-08-20

    申请号：US15691886

    申请日：2017-08-31

    Applicant: Tsinghua University ,  HON HAI PRECISION INDUSTRY CO., LTD.

    Inventor： Jun Zhu ,  Wei Hou ,  Guo-Fan Jin ,  Shou-Shan Fan

    IPC: G02B17/06 ,  G02B27/00

    Abstract: A oblique camera lens includes: a primary mirror configured to reflect a light ray to form a first reflected light; a secondary mirror located on a first path of light reflected from the primary mirror and configured to reflect the first reflected light to form a second reflected light; a tertiary mirror located on a second path of light reflected from the secondary mirror and configured to reflect the second reflected light to form a third reflected light; and an image sensor located on a third path of light reflected from the tertiary mirror and configured to receive the third reflected light; wherein each of the first reflecting surface and the third reflecting surface is a sixth order xy polynomial freeform surface; and a field of view of oblique camera lens in an Y-axis direction is greater or equal to 35° and less than or equal to 65°.

Patent Agency Ranking