中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

16 records (took 0.026 seconds)

    Introspection network for training neural networks
    11.
    发明授权
    Introspection network for training neural networks 有权

    公开(公告)号：US10755199B2

    公开(公告)日：2020-08-25

    申请号：US15608517

    申请日：2017-05-30

    Applicant: Adobe Inc.

    Inventor： Mausoom Sarkar ,  Balaji Krishnamurthy ,  Abhishek Sinha ,  Aahitagni Mukherjee

    IPC: G06N20/00 ,  G06F7/02 ,  G06N3/08 ,  G06N3/04

    Abstract: An introspection network is a machine-learned neural network that accelerates training of other neural networks. The introspection network receives a weight history for each of a plurality of weights from a current training step for a target neural network. A weight history includes at least four values for the weight that are obtained during training of the target neural network up to the current step. The introspection network then provides, for each of the plurality of weights, a respective predicted value, based on the weight history. The predicted value for a weight represents a value for the weight in a future training step for the target neural network. Thus, the predicted value represents a jump in the training steps of the target neural network, which reduces the training time of the target neural network. The introspection network then sets each of the plurality of weights to its respective predicted value.

    Systems and methods of training neural networks against adversarial attacks
    13.
    发明授权
    Systems and methods of training neural networks against adversarial attacks 有权

    公开(公告)号：US11468314B1

    公开(公告)日：2022-10-11

    申请号：US16129553

    申请日：2018-09-12

    Applicant: ADOBE INC.

    Inventor： Mayank Singh ,  Abhishek Sinha ,  Balaji Krishnamurthy

    IPC: G06N3/08 ,  G06K9/62 ,  G06N3/04 ,  G06F21/57 ,  G06N5/04

    Abstract: Embodiments disclosed herein describe systems, methods, and products that generate trained neural networks that are robust against adversarial attacks. During a training phase, an illustrative computer may iteratively optimize a loss function that may include a penalty for ill-conditioned weight matrices in addition to a penalty for classification errors. Therefore, after the training phase, the trained neural network may include one or more well-conditioned weight matrices. The one or more well-conditioned weight matrices may minimize the effect of perturbations within an adversarial input thereby increasing the accuracy of classification of the adversarial input. By contrast, conventional training approaches may merely reduce the classification errors using backpropagation, and, as a result, any perturbation in an input is prone to generate a large effect on the output.

    IDENTIFYING DIGITAL ATTRIBUTES FROM MULTIPLE ATTRIBUTE GROUPS UTILIZING A DEEP COGNITIVE ATTRIBUTION NEURAL NETWORK
    14.
    发明申请
    IDENTIFYING DIGITAL ATTRIBUTES FROM MULTIPLE ATTRIBUTE GROUPS UTILIZING A DEEP COGNITIVE ATTRIBUTION NEURAL NETWORK 有权

    公开(公告)号：US20220309093A1

    公开(公告)日：2022-09-29

    申请号：US17806922

    申请日：2022-06-14

    Applicant: Adobe Inc.

    Inventor： Ayush Chopra ,  Mausoom Sarkar ,  Jonas Dahl ,  Hiresh Gupta ,  Balaji Krishnamurthy ,  Abhishek Sinha

    IPC: G06F16/535 ,  G06K9/62 ,  G06F17/15 ,  G06N3/04 ,  G06F16/55

    Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media for generating tags for an object portrayed in a digital image based on predicted attributes of the object. For example, the disclosed systems can utilize interleaved neural network layers of alternating inception layers and dilated convolution layers to generate a localization feature vector. Based on the localization feature vector, the disclosed systems can generate attribute localization feature embeddings, for example, using some pooling layer such as a global average pooling layer. The disclosed systems can then apply the attribute localization feature embeddings to corresponding attribute group classifiers to generate tags based on predicted attributes. In particular, attribute group classifiers can predict attributes as associated with a query image (e.g., based on a scoring comparison with other potential attributes of an attribute group). Based on the generated tags, the disclosed systems can respond to tag queries and search queries.

    Digital image search training using aggregated digital images
    16.
    发明授权
    Digital image search training using aggregated digital images 有权

    公开(公告)号：US10831818B2

    公开(公告)日：2020-11-10

    申请号：US16177243

    申请日：2018-10-31

    Applicant: Adobe Inc.

    Inventor： Mausoom Sarkar ,  Hiresh Gupta ,  Abhishek Sinha

    IPC: G06K9/54 ,  G06F16/58 ,  G06N3/08 ,  G06F16/56 ,  G06F16/583

    Abstract: Digital image search training techniques and machine-learning architectures are described. In one example, a query digital image is received by service provider system, which is then used to select at least one positive sample digital image, e.g., having a same product ID. A plurality of negative sample digital images is also selected by the service provider system based on the query digital image, e.g., having different product IDs. The at least one positive sample digital image and the plurality of negative samples are then aggregated by the service provider system into a single aggregated digital image. At least one neural network is then trained by the service provider system using a loss function based on a feature comparison between the query digital image and samples from the aggregated digital image in a single pass.

Patent Agency Ranking