公开(公告)号：US20250094838A1

公开(公告)日：2025-03-20

申请号：US18967327

申请日：2024-12-03

Applicant: Google LLC

Inventor： Jason Weng Wei ,  Dengyong Zhou ,  Xuezhi Wang ,  Dale Eric Schuurmans ,  Quoc V. Le ,  Maarten Paul Bosma ,  Ed Huai-Hsin Chi ,  Olivier Jean Andrè Bousquet ,  Le Hou ,  Charles Aloysius Sutton ,  Nathanael Martin Schärli ,  Nathan Kemp Sekiguchi Scales ,  Augustus Quadrozzi Odena ,  Sharan Ajit Narang ,  Guy Gur-Ari Krakover ,  Aakanksha Chowdhery ,  David Martin Dohan ,  Aitor Lewkowycz ,  Jacob Austin ,  Henryk Michalewski ,  David Luan ,  David J. Bieber ,  Anders Johan Andreassen ,  Maxwell Isaac Nye

IPC: G06N5/022

Abstract: An example technique for image analysis is provided. An example image analysis method includes obtaining an instructive sequence descriptive of an instructive query, an instructive response, and an instructive trace of intermediate states from the instructive query to the instructive response. The example image analysis method includes inputting, to a machine-learned model, the instructive sequence and an operative image processing query that comprises image data, wherein the machine-learned model is configured to process the operative query with attention over the instructive sequence. The example method can include generating, using the machine-learned model and responsive to the operative query, an operative image processing response that comprises an analysis of the image data.

公开(公告)号：US11514313B2

公开(公告)日：2022-11-29

申请号：US16580649

申请日：2019-09-24

Applicant: Google LLC

Inventor： Samaneh Azadi ,  Ian Goodfellow ,  Catherine Olsson ,  Augustus Quadrozzi Odena

IPC: G06N3/08 ,  G06N3/04 ,  G10L15/16

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing a data sample in response to a request for a data sample. In one aspect, a method comprises: receiving a request for a new data sample; until a candidate new data sample is generated that satisfies an acceptance criterion, performing operations comprising: generating a candidate new data sample using a generator neural network; processing the candidate new data sample using a discriminator neural network to generate an imitation score; and determining, from the imitation score, whether the candidate new data sample satisfies the acceptance criterion; and providing the candidate new data sample that satisfies the acceptance criterion in response to the received request.

公开(公告)号：US11080603B2

公开(公告)日：2021-08-03

申请号：US16415693

申请日：2019-05-17

Applicant: Google LLC

Inventor： Augustus Quadrozzi Odena

IPC: G06N3/10 ,  G06F11/36 ,  G06N3/08

Abstract: The present disclosure provides systems and methods for debugging neural networks. In one example, a computer-implemented method is provided, which includes obtaining, by one or more computing devices, one or more inputs from an input corpus. The method further includes mutating, by the one or more computing devices, the one or more inputs and providing the one or more mutated inputs to a neural network; obtaining, by the one or more computing devices as a result of the neural network processing the one or more mutated inputs, a set of coverage arrays; determining, by the one or more computing devices based at least in part on the set of coverage arrays, whether the one or more mutated inputs provide new coverage; and upon determining that the one or more mutated inputs provide new coverage, adding the one or more mutated inputs to the input corpus.

公开(公告)号：US11790211B2

公开(公告)日：2023-10-17

申请号：US15884253

申请日：2018-01-30

Applicant: Google LLC

Inventor： Augustus Quadrozzi Odena ,  John Dieterich Lawson

IPC: G06N3/045 ,  G06N3/08 ,  G06N3/044

CPC classification number: G06N3/045 ,  G06N3/044 ,  G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for adjusting neural network resource usage. One of the methods includes receiving a network input for processing by a task neural network, the task neural network comprising a plurality of neural network layers; receiving a usage input specifying a respective weight for each of one or more usage factors, wherein each usage factor impacts how many computational resources are used by the task neural network during the processing of the network input; and processing the network input using the task neural network in accordance with the usage input to generate a network output for the network input, comprising: selecting, based at least on the usage input, a proper subset of the plurality of neural network layers to be active while processing the network input, and processing the network input using only the selected neural network layers.

公开(公告)号：US20210365797A1

公开(公告)日：2021-11-25

申请号：US17392937

申请日：2021-08-03

Applicant: Google LLC

Inventor： Augustus Quadrozzi Odena

IPC: G06N3/10 ,  G06F11/36 ,  G06N3/08

Abstract: The present disclosure provides systems and methods for debugging neural networks. In one example, a computer-implemented method is provided, which includes obtaining, by one or more computing devices, one or more inputs from an input corpus. The method further includes mutating, by the one or more computing devices, the one or more inputs and providing the one or more mutated inputs to a neural network; obtaining, by the one or more computing devices as a result of the neural network processing the one or more mutated inputs, a set of coverage arrays; determining, by the one or more computing devices based at least in part on the set of coverage arrays, whether the one or more mutated inputs provide new coverage; and upon determining that the one or more mutated inputs provide new coverage, adding the one or more mutated inputs to the input corpus.

公开(公告)号：US20190354870A1

公开(公告)日：2019-11-21

申请号：US16415693

申请日：2019-05-17

Applicant: Google LLC

Inventor： Augustus Quadrozzi Odena

IPC: G06N3/10 ,  G06N3/08 ,  G06F11/36

Abstract: The present disclosure provides systems and methods for debugging neural networks. In one example, a computer-implemented method is provided, which includes obtaining, by one or more computing devices, one or more inputs from an input corpus. The method further includes mutating, by the one or more computing devices, the one or more inputs and providing the one or more mutated inputs to a neural network; obtaining, by the one or more computing devices as a result of the neural network processing the one or more mutated inputs, a set of coverage arrays; determining, by the one or more computing devices based at least in part on the set of coverage arrays, whether the one or more mutated inputs provide new coverage; and upon determining that the one or more mutated inputs provide new coverage, adding the one or more mutated inputs to the input corpus.

公开(公告)号：US20190236438A1

公开(公告)日：2019-08-01

申请号：US15884253

申请日：2018-01-30

Applicant: Google LLC

Inventor： Augustus Quadrozzi Odena ,  John Dieterich Lawson

IPC: G06N3/04 ,  G06N3/08

CPC classification number: G06N3/0454 ,  G06N3/0445 ,  G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for adjusting neural network resource usage. One of the methods includes receiving a network input for processing by a task neural network, the task neural network comprising a plurality of neural network layers; receiving a usage input specifying a respective weight for each of one or more usage factors, wherein each usage factor impacts how many computational resources are used by the task neural network during the processing of the network input; and processing the network input using the task neural network in accordance with the usage input to generate a network output for the network input, comprising: selecting, based at least on the usage input, a proper subset of the plurality of neural network layers to be active while processing the network input, and processing the network input using only the selected neural network layers.

公开(公告)号：US20240185030A1

公开(公告)日：2024-06-06

申请号：US18487802

申请日：2023-10-16

Applicant: Google LLC

Inventor： Augustus Quadrozzi Odena ,  John Dieterich Lawson

IPC: G06N3/045 ,  G06N3/044 ,  G06N3/08

CPC classification number: G06N3/045 ,  G06N3/044 ,  G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for adjusting neural network resource usage. One of the methods includes receiving a network input for processing by a task neural network, the task neural network comprising a plurality of neural network layers; receiving a usage input specifying a respective weight for each of one or more usage factors, wherein each usage factor impacts how many computational resources are used by the task neural network during the processing of the network input; and processing the network input using the task neural network in accordance with the usage input to generate a network output for the network input, comprising: selecting, based at least on the usage input, a proper subset of the plurality of neural network layers to be active while processing the network input, and processing the network input using only the selected neural network layers.

公开(公告)号：US20230394328A1

公开(公告)日：2023-12-07

申请号：US17881746

申请日：2022-08-05

Applicant: Google LLC

Inventor： Jason Weng Wei ,  Dengyong Zhou ,  Dale Eric Schuurmans ,  Quoc V. Le ,  Maarten Paul Bosma ,  Ed Huai-Hsin Chi ,  Olivier Jean Andrè Bousquet ,  Le Hou ,  Nathan Kemp Sekiguchi Scales ,  David J. Bieber ,  Charles Aloysius Sutton ,  Nathanael Martin Schärli ,  Augustus Quadrozzi Odena ,  Sharan Ajit Narang ,  Guy Gur-Ari Krakover ,  Aakanksha Chowdhery ,  Aitor Lewkowycz ,  Jiageng Luan ,  David Martin Dohan ,  Henryk Michalewski ,  Jacob Austin ,  Anders Johan Andreassen ,  Maxwell Isaac Nye ,  Xuezhi Wang

IPC: G06N5/02

CPC classification number: G06N5/022

Abstract: Example embodiments of aspects of the present disclosure provide an example computer-implemented method for improved prompting of a machine-learned model. The example method can include obtaining an instructive sequence descriptive of an instructive query, an instructive response, and an instructive trace of intermediate states from the instructive query to the instructive response. The example method can include inputting, to a machine-learned model, the instructive sequence and an operative query, wherein the machine-learned model is configured to process the operative query with attention over the instructive sequence. The example method can include generating, using the machine-learned model and responsive to the operative query, an operative response.

公开(公告)号：US20230244938A1

公开(公告)日：2023-08-03

申请号：US18160776

申请日：2023-01-27

Applicant: Google LLC

Inventor： Jason Weng Wei ,  Dengyong Zhou ,  Xuezhi Wang ,  Dale Eric Schuurmans ,  Quoc V. Le ,  Maarten Paul Bosma ,  Ed Huai-Hsin Chi ,  Olivier Jean Andrè Bousquet ,  Le Hou ,  Charles Aloysius Sutton ,  Nathanael Martin Schärli ,  Nathan Kemp Sekiguchi Scales ,  Augustus Quadrozzi Odena ,  Sharan Ajit Narang ,  Guy Gur-Ari Krakover ,  Aakanksha Chowdhery ,  David Martin Dohan ,  Aitor Lewkowycz ,  Henryk Michalewski ,  Jiageng Luan ,  David J. Bieber ,  Jacob Austin ,  Anders Johan Andreassen ,  Maxwell Isaac Nye ,  Yi Tay ,  Mostafa Dehghani

IPC: G06N3/08

CPC classification number: G06N3/08

Abstract: An example method for pretraining a machine-learned model is provided. The example method includes obtaining a plurality of different combinations of configuration parameters of a pretraining objective framework. The example method includes generating, using the pretraining objective framework, a plurality of corrupted training examples from one or more training examples, wherein the plurality of corrupted training examples are respectively generated according to the plurality of different combinations. The example method includes inputting the plurality of corrupted training examples into the machine-learned model, wherein the machine-learned model is configured to generate uncorrupted subportions corresponding to corrupted subportions of the corrupted training examples. The example method includes obtaining, from the machine-learned model, a plurality of outputs respectively generated by the machine-learned model based on the plurality of corrupted training examples. The example method includes updating one or more parameters of the machine-learned model based on an evaluation of the plurality of outputs.