Patents by Inventor Gregory S. Corrado
Gregory S. Corrado has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11954597Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.Type: GrantFiled: October 24, 2022Date of Patent: April 9, 2024Assignee: Google LLCInventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Gary R. Holt, Julian P. Grady, Sharat Chikkerur, David W. Sculley, II
-
Publication number: 20240070392Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing numeric representations of words. One of the methods includes obtaining a set of training data, wherein the set of training data comprises sequences of words; training a classifier and an embedding function on the set of training data, wherein training the embedding function comprises obtained trained values of the embedding function parameters; processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary in the high-dimensional space; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.Type: ApplicationFiled: November 6, 2023Publication date: February 29, 2024Inventors: Tomas Mikolov, Kai Chen, Gregory S. Corrado, Jeffrey A. Dean
-
Patent number: 11809824Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing numeric representations of words. One of the methods includes obtaining a set of training data, wherein the set of training data comprises sequences of words; training a classifier and an embedding function on the set of training data, wherein training the embedding function comprises obtained trained values of the embedding function parameters; processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary in the high-dimensional space; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.Type: GrantFiled: February 12, 2021Date of Patent: November 7, 2023Assignee: Google LLCInventors: Tomas Mikolov, Kai Chen, Gregory S. Corrado, Jeffrey A. Dean
-
Publication number: 20230325657Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.Type: ApplicationFiled: October 24, 2022Publication date: October 12, 2023Inventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Gary R. Holt, Julian P. Grady, Sharat Chikkerur, David W. Sculley, II
-
Patent number: 11687832Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.Type: GrantFiled: August 3, 2020Date of Patent: June 27, 2023Assignee: Google LLCInventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Samy Bengio, Rajat Monga, Matthieu Devin
-
Patent number: 11481631Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.Type: GrantFiled: June 8, 2020Date of Patent: October 25, 2022Assignee: Google LLCInventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Gary R. Holt, Julian P Grady, Sharat Chikkerur, David W. Sculley, II
-
Patent number: 10922488Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing numeric representations of words. One of the methods includes obtaining a set of training data, wherein the set of training data comprises sequences of words; training a classifier and an embedding function on the set of training data, wherein training the embedding function comprises obtained trained values of the embedding function parameters; processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary in the high-dimensional space; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.Type: GrantFiled: March 25, 2019Date of Patent: February 16, 2021Assignee: Google LLCInventors: Tomas Mikolov, Kai Chen, Gregory S. Corrado, Jeffrey A. Dean
-
Patent number: 10733535Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.Type: GrantFiled: July 31, 2017Date of Patent: August 4, 2020Assignee: Google LLCInventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Samy Bengio, Rajat Monga, Matthieu Devin
-
Patent number: 10679124Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.Type: GrantFiled: December 2, 2016Date of Patent: June 9, 2020Assignee: Google LLCInventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Gary R. Holt, Julian P. Grady, Sharat Chikkerur, David W. Sculley, II
-
Patent number: 10241997Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing numeric representations of words. One of the methods includes obtaining a set of training data, wherein the set of training data comprises sequences of words; training a classifier and an embedding function on the set of training data, wherein training the embedding function comprises obtained trained values of the embedding function parameters; processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary in the high-dimensional space; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.Type: GrantFiled: August 21, 2017Date of Patent: March 26, 2019Assignee: Google LLCInventors: Tomas Mikolov, Kai Chen, Gregory S. Corrado, Jeffrey A. Dean
-
Patent number: 10127475Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying images.Type: GrantFiled: September 22, 2016Date of Patent: November 13, 2018Assignee: Google LLCInventors: Gregory S. Corrado, Jeffrey A. Dean, Samy Bengio, Andrea L. Frome, Jonathon Shlens
-
Patent number: 9740680Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing numeric representations of words. One of the methods includes obtaining a set of training data, wherein the set of training data comprises sequences of words; training a classifier and an embedding function on the set of training data, wherein training the embedding function comprises obtained trained values of the embedding function parameters; processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary in the high-dimensional space; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.Type: GrantFiled: May 18, 2015Date of Patent: August 22, 2017Assignee: Google Inc.Inventors: Tomas Mikolov, Kai Chen, Gregory S. Corrado, Jeffrey A. Dean
-
Patent number: 9721214Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.Type: GrantFiled: August 8, 2016Date of Patent: August 1, 2017Assignee: Google Inc.Inventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Samy Bengio, Rajat Monga, Matthieu Devin
-
Patent number: 9514404Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.Type: GrantFiled: September 21, 2015Date of Patent: December 6, 2016Assignee: Google Inc.Inventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Gary R. Holt, Julian P. Grady, Sharat Chikkerur, David W. Sculley, II
-
Patent number: 9514405Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for scoring concept terms using a deep network. One of the methods includes receiving an input comprising a plurality of features of a resource, wherein each feature is a value of a respective attribute of the resource; processing each of the features using a respective embedding function to generate one or more numeric values; processing the numeric values to generate an alternative representation of the features of the resource, wherein processing the floating point values comprises applying one or more non-linear transformations to the floating point values; and processing the alternative representation of the input to generate a respective relevance score for each concept term in a pre-determined set of concept terms, wherein each of the respective relevance scores measures a predicted relevance of the corresponding concept term to the resource.Type: GrantFiled: September 21, 2015Date of Patent: December 6, 2016Assignee: Google Inc.Inventors: Kai Chen, Xiaodan Song, Gregory S. Corrado, Kun Zhang, Jeffrey A. Dean, Bahman Rabii
-
Patent number: 9412065Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.Type: GrantFiled: August 4, 2015Date of Patent: August 9, 2016Assignee: Google Inc.Inventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Samy Bengio, Rajat Monga, Matthieu Devin
-
Patent number: 9256807Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating labeled images. One of the methods includes selecting a plurality of candidate videos from videos identified in a response to a search query derived from a label for an object category; selecting one or more initial frames from each of the candidate videos; detecting one or more initial images of objects in the object category in the initial frames; for each initial frame including an initial image of an object in the object category, tracking the object through surrounding frames to identify additional images of the object; and selecting one or more images from the one or more initial images and one or more additional images as database images of objects belonging to the object category.Type: GrantFiled: March 14, 2013Date of Patent: February 9, 2016Assignee: Google Inc.Inventors: Jonathon Shlens, Quoc V. Le, Gregory S. Corrado, Marc'Aurelio Ranzato
-
Publication number: 20160012331Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for scoring concept terms using a deep network. One of the methods includes receiving an input comprising a plurality of features of a resource, wherein each feature is a value of a respective attribute of the resource; processing each of the features using a respective embedding function to generate one or more numeric values; processing the numeric values to generate an alternative representation of the features of the resource, wherein processing the floating point values comprises applying one or more non-linear transformations to the floating point values; and processing the alternative representation of the input to generate a respective relevance score for each concept term in a pre-determined set of concept terms, wherein each of the respective relevance scores measures a predicted relevance of the corresponding concept term to the resource.Type: ApplicationFiled: September 21, 2015Publication date: January 14, 2016Inventors: Kai Chen, Xiaodan Song, Gregory S. Corrado, Kun Zhang, Jeffrey A. Dean, Bahman Rabii
-
Patent number: 9218573Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.Type: GrantFiled: March 14, 2013Date of Patent: December 22, 2015Assignee: Google Inc.Inventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Samy Bengio, Rajat Monga, Matthieu Devin
-
Patent number: 9141916Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.Type: GrantFiled: March 14, 2013Date of Patent: September 22, 2015Assignee: Google Inc.Inventors: Gregory S. Corrado, Kai Chen, Jeffrey A. Dean, Gary R. Holt, Julian P. Grady, Sharat Chikkerur, David W. Sculley