公开(公告)号：US20250005798A1

公开(公告)日：2025-01-02

申请号：US18883946

申请日：2024-09-12

Applicant: Google LLC

Inventor： Neil Matthew Tinmouth Houlsby ,  Sylvain Gelly ,  Jakob D. Uszkoreit ,  Xiaohua Zhai ,  Georg Heigold ,  Lucas Klaus Beyer ,  Alexander Kolesnikov ,  Matthias Johannes Lorenz Minderer ,  Dirk Weissenborn ,  Mostafa Dehghani ,  Alexey Dosovitskiy ,  Thomas Unterthiner

IPC: G06T7/00 ,  G06F18/24 ,  G06N3/045 ,  G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.

公开(公告)号：US20250005797A1

公开(公告)日：2025-01-02

申请号：US18883917

申请日：2024-09-12

Applicant: Google LLC

Inventor： Neil Matthew Tinmouth Houlsby ,  Sylvain Gelly ,  Jakob D. Uszkoreit ,  Xiaohua Zhai ,  Georg Heigold ,  Lucas Klaus Beyer ,  Alexander Kolesnikov ,  Matthias Johannes Lorenz Minderer ,  Dirk Weissenborn ,  Mostafa Dehghani ,  Alexey Dosovitskiy ,  Thomas Unterthiner

IPC: G06T7/00 ,  G06F18/24 ,  G06N3/045 ,  G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.

公开(公告)号：US20240428587A1

公开(公告)日：2024-12-26

申请号：US18827133

申请日：2024-09-06

Applicant: Google LLC

Inventor： Anurag Arnab ,  Mostafa Dehghani ,  Georg Heigold ,  Chen Sun ,  Mario Lucic ,  Cordelia Luise Schmid

IPC: G06V20/40 ,  G06N20/00

Abstract: A computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of video tokens from the video data, the plurality of video tokens comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of video tokens as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.

公开(公告)号：US12112538B2

公开(公告)日：2024-10-08

申请号：US17370522

申请日：2021-07-08

Applicant: Google LLC

Inventor： Anurag Arnab ,  Mostafa Dehghani ,  Georg Heigold ,  Chen Sun ,  Mario Lucic ,  Cordelia Luise Schmid

IPC: G06V20/40 ,  G06N20/00

CPC classification number: G06V20/41 ,  G06N20/00 ,  G06V20/46 ,  G06V20/49

Abstract: A computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of video tokens from the video data, the plurality of video tokens comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of video tokens as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.

公开(公告)号：US11886976B1

公开(公告)日：2024-01-30

申请号：US18222395

申请日：2023-07-14

Applicant: Google LLC

Inventor： Tal Schuster ,  Adam Joshua Fisch ,  Jai Prakash Gupta ,  Mostafa Dehghani ,  Dara Bahri ,  Vinh Quoc Tran ,  Yi Tay ,  Donald Arthur Metzler, Jr.

IPC: G06N3/0455

CPC classification number: G06N3/0455

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output sequences using auto-regressive decoder neural networks. In particular, during generation, adaptive early exiting is used to reduce the time required to generate the output sequence.

公开(公告)号：US20230017072A1

公开(公告)日：2023-01-19

申请号：US17370522

申请日：2021-07-08

Applicant: Google LLC

Inventor： Anurag Arnab ,  Mostafa Dehghani ,  Georg Heigold ,  Chen Sun ,  Mario Lucic ,  Cordelia Luise Schmid

IPC: G06K9/00 ,  G06N20/00

Abstract: A computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of video tokens from the video data, the plurality of video tokens comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of video tokens as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.

公开(公告)号：US20240403636A1

公开(公告)日：2024-12-05

申请号：US18697257

申请日：2022-10-05

Applicant: GOOGLE LLC

Inventor： Valerii Likhosherstov ,  Mostafa Dehghani ,  Anurag Arnab ,  Krzysztof Marcin Choromanski ,  Mario Lucic ,  Yi Tay

IPC: G06N3/08 ,  G06N3/045

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for executing and training a multi-modal, multi-task self-attention neural network.

公开(公告)号：US20240256964A1

公开(公告)日：2024-08-01

申请号：US18424031

申请日：2024-01-26

Applicant: Google LLC

Inventor： Yi Tay ,  Mostafa Dehghani

IPC: G06N20/00 ,  G06F7/483

CPC classification number: G06N20/00 ,  G06F7/483

Abstract: An example method includes obtaining a pretrained machine-learned model that was initially pretrained using a pretraining dataset and further pretraining the model by generating, using a pretraining objective framework, a plurality of corrupted training examples from one or more training examples obtained from the pretraining dataset. A first set of one or more training examples can be corrupted according to a first set of configuration parameters of the pretraining objective framework. A second set can be corrupted according to a second set of configuration parameters of the pretraining objective framework. The example method includes inputting the plurality of corrupted training examples into model; obtaining from the model, a plurality of outputs respectively generated by model based on the plurality of corrupted training examples; and updating one or more parameters of model based on an evaluation of the plurality of outputs.

公开(公告)号：US11983903B2

公开(公告)日：2024-05-14

申请号：US18500034

申请日：2023-11-01

Applicant: Google LLC

Inventor： Neil Matthew Tinmouth Houlsby ,  Sylvain Gelly ,  Jakob D. Uszkoreit ,  Xiaohua Zhai ,  Georg Heigold ,  Lucas Klaus Beyer ,  Alexander Kolesnikov ,  Matthias Johannes Lorenz Minderer ,  Dirk Weissenborn ,  Mostafa Dehghani ,  Alexey Dosovitskiy ,  Thomas Unterthiner

IPC: G06T7/00 ,  G06F18/24 ,  G06N3/045 ,  G06N3/08

CPC classification number: G06T7/97 ,  G06F18/24 ,  G06N3/045 ,  G06N3/08 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.

公开(公告)号：US20220245432A1

公开(公告)日：2022-08-04

申请号：US17592174

申请日：2022-02-03

Applicant: Google LLC

Inventor： Yi Tay ,  Donald Arthur Metzler, JR. ,  Dara Bahri ,  Mostafa Dehghani

IPC: G06N3/04 ,  G06F40/20

Abstract: The present disclosure provides echo-attention layers, a new efficient method for increasing the expressiveness of self-attention layers without incurring significant parameter or training time costs. One intuition behind the proposed method is to learn to echo, i.e., attend once and then get N echo-ed attentions for free (or at a relatively cheap cost). As compared to stacking new layers, the proposed echoed attentions are targeted at providing similar representation power at a better cost efficiency.