Patents by Inventor Stuart Ozer
Stuart Ozer 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: 11893462Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for sharing, on a distributed database, a database application to a first user of the distributed database, the database application generated by a second user of the distributed database. The training dataset includes a first database training dataset from the first user of the distributed database and a second database training dataset from the second user of the distributed database, the first database training dataset and the second database training dataset including non-overlapping dataset features. The database application further identifies a query from the second user to train the machine learning model on the training dataset and generates a trained machine learning model by training the machine learning model on a joined dataset according to the query. The database application generates outputs from the trained machine learning model by applying the trained machine learning model on new data.Type: GrantFiled: November 14, 2022Date of Patent: February 6, 2024Assignee: Snowflake Inc.Inventors: Monica J. Holboke, Justin Langseth, Stuart Ozer, William L. Stratton, Jr.
-
Publication number: 20230409589Abstract: The subject technology generates, by a database system, cell data for a particular table based on values from a source table, the values being based on raw input data, the source table comprising multiple rows and multiple columns, the raw input data comprising values in a first format, the values comprising input features corresponding to datasets included in the raw input data for machine learning models, the source table being provided by an external environment, the external environment comprising an external system from the database system. The subject technology performs a database operation to generate the particular table including table metadata, column metadata, and the generated cell data, the generated particular table comprising a second format that causes more efficient processing of data by the database system using a single query on the particular table compared to processing the raw input data from the source table.Type: ApplicationFiled: August 30, 2023Publication date: December 21, 2023Inventors: Simon A. Field, Stuart Ozer
-
Patent number: 11775544Abstract: The subject technology receives by a database system, raw input data from a source table provided by an external environment, the source table comprising multiple rows and multiple columns, the raw input data comprising values in a first format, the values comprising input features corresponding to datasets included in the raw input data for machine learning models, the external environment comprising an external system from the database system and is accessed by different users. The subject technology generates cell data for a second table based on the values from the source table. The subject technology performs a database operation to generate the second table including table metadata, column metadata, and the generated cell data.Type: GrantFiled: January 31, 2023Date of Patent: October 3, 2023Assignee: Snowflake Inc.Inventors: Simon A. Field, Stuart Ozer
-
Publication number: 20230186160Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for sharing, on a distributed database, a database application to a first user of the distributed database, the database application generated by a second user of the distributed database. The training dataset includes a first database training dataset from the first user of the distributed database and a second database training dataset from the second user of the distributed database, the first database training dataset and the second database training dataset including non-overlapping dataset features. The database application further identifies a query from the second user to train the machine learning model on the training dataset and generates a trained machine learning model by training the machine learning model on a joined dataset according to the query. The database application generates outputs from the trained machine learning model by applying the trained machine learning model on new data.Type: ApplicationFiled: November 14, 2022Publication date: June 15, 2023Inventors: Monica J. Holboke, Justin Langseth, Stuart Ozer, William L. Stratton, JR.
-
Publication number: 20230177063Abstract: The subject technology receives by a database system, raw input data from a source table provided by an external environment, the source table comprising multiple rows and multiple columns, the raw input data comprising values in a first format, the values comprising input features corresponding to datasets included in the raw input data for machine learning models, the external environment comprising an external system from the database system and is accessed by different users. The subject technology generates cell data for a second table based on the values from the source table. The subject technology performs a database operation to generate the second table including table metadata, column metadata, and the generated cell data.Type: ApplicationFiled: January 31, 2023Publication date: June 8, 2023Inventors: Simon A. Field, Stuart Ozer
-
Publication number: 20230169407Abstract: A system for providing access to a database management system (DBMS) to a first user of a cloud data platform, the DBMS being generated by a second user. A machine learning model for training on a training dataset is included in the DBMS. The training dataset includes a first training dataset that is encrypted in the DBMS and a second training dataset that includes non-overlapping features with the first training dataset. A request, from the second user, to train the machine learning model on the first and second training datasets is identified. A trained machine learning model is generated by training the machine learning model on a joined dataset according to the request. One or more outputs from the trained machine learning model are generated by applying the trained machine learning model on new data.Type: ApplicationFiled: January 31, 2023Publication date: June 1, 2023Inventors: Monica J. Holboke, Justin Langseth, Stuart Ozer, William L. Stratton, JR.
-
Patent number: 11609927Abstract: The subject technology receives, by a database system, raw input data from a source table provided by a machine learning development environment, the source table comprising multiple rows where each row includes multiple columns, the raw input data comprising values in a first format, the values comprising input features corresponding to datasets included in the raw input data for machine learning models, the machine learning development environment comprising an external system from the database system and is accessed by a plurality of different users that are external to the database system. The subject technology generates cell data for a feature store table based at least in part on the values from the source table. The subject technology performs at least one database operation to generate the feature store table including at least table metadata, column metadata, and the generated cell data.Type: GrantFiled: August 30, 2022Date of Patent: March 21, 2023Assignee: Snowflake Inc.Inventors: Simon A. Field, Stuart Ozer
-
Publication number: 20230004571Abstract: The subject technology receives, by a database system, raw input data from a source table provided by a machine learning development environment, the source table comprising multiple rows where each row includes multiple columns, the raw input data comprising values in a first format, the values comprising input features corresponding to datasets included in the raw input data for machine learning models, the machine learning development environment comprising an external system from the database system and is accessed by a plurality of different users that are external to the database system. The subject technology generates cell data for a feature store table based at least in part on the values from the source table. The subject technology performs at least one database operation to generate the feature store table including at least table metadata, column metadata, and the generated cell data.Type: ApplicationFiled: August 30, 2022Publication date: January 5, 2023Inventors: Simon A. Field, Stuart Ozer
-
Patent number: 11501015Abstract: A secure machine learning system of a database system can be implemented to use secure shared data to train a machine learning model. To manage the model, a first user of the database can share data in an encrypted view with a second user of the database, and further share one or more functions of an application that accesses the data while the data is encrypted. The second user can access functions of the application and can call the functions to generate a trained machine learning model and further generate machine learning outputs (e.g., predictions) from the trained model.Type: GrantFiled: December 16, 2021Date of Patent: November 15, 2022Assignee: Snowflake Inc.Inventors: Monica J. Holboke, Justin Langseth, Stuart Ozer, William L. Stratton, Jr.
-
Patent number: 11461351Abstract: The subject technology receives raw input data from a source table, the raw input data including data comprising input features for a machine learning model, the raw input data being in a first format including at least multiple rows with each row including multiple columns of values. Based at least in part on the source table, the subject technology generates table metadata corresponding to the source table. Based at least in part on the received raw input data, the subject technology generates column metadata corresponding to values from the source table. The subject technology generates cell data for a feature store table based at least in part on the values from the source table. The subject technology performs at least one database operation to generate the feature store table including at least the generated table metadata, the generated column metadata, and the generated cell data.Type: GrantFiled: July 31, 2021Date of Patent: October 4, 2022Assignee: Snowflake Inc.Inventors: Simon A. Field, Stuart Ozer
-
Publication number: 20220292213Abstract: A secure machine learning system of a database system can be implemented to use secure shared data to train a machine learning model. To manage the model, a first user of the database can share data in an encrypted view with a second user of the database, and further share one or more functions of an application that accesses the data while the data is encrypted. The second user can access functions of the application and can call the functions to generate a trained machine learning model and further generate machine learning outputs (e.g., predictions) from the trained model.Type: ApplicationFiled: December 16, 2021Publication date: September 15, 2022Inventors: Monica J. Holboke, Justin Langseth, Stuart Ozer, William L. Stratton, JR.
-
Patent number: 11216580Abstract: A secure machine learning system of a database system can be implemented to use secure shared data to train a machine learning model. To manage the model, a first user of the database can share data in an encrypted view with a second user of the database, and further share one or more functions of an application that accesses the data while the data is encrypted. The second user can access functions of the application and can call the functions to generate a trained machine learning model and further generate machine learning outputs (e.g., predictions) from the trained model.Type: GrantFiled: April 16, 2021Date of Patent: January 4, 2022Assignee: Snowflake Inc.Inventors: Monica J. Holboke, Justin Langseth, Stuart Ozer, William L. Stratton, Jr.
-
Patent number: 8639445Abstract: Data clustering is described for determining related components of the data. In one example, information on biomolecules may be clustered into groups in which the biomolecular data in a clustered group may indicate data that shares a relationship. For example, monomers of a protein molecule or a nucleic acid molecule may be mapped to an evolutionary or phylogenetic tree. Candidate groupings of the information may be obtained based on evolutionary relationships among sequences corresponding to the molecules.Type: GrantFiled: July 23, 2007Date of Patent: January 28, 2014Assignee: Microsoft CorporationInventor: Stuart Ozer
-
Patent number: 8001567Abstract: A media planner displays descriptors that represent programs that are scheduled for broadcast on tiles according to a layout. The layout arranges the tiles according to a day and a day part based on a scheduled broadcast date and time associated with a program.Type: GrantFiled: May 2, 2002Date of Patent: August 16, 2011Assignee: Microsoft CorporationInventors: Stuart Ozer, Wei Wei Ada Cho, Warren Neal Thornthwaite
-
Patent number: 7870023Abstract: A system, including a planning module, a control module and a receiver module, configured to schedule display of one or more advertising impressions of available advertising inventory. The planning module enables scheduling a requested quantity of advertising impressions in accordance with target criteria. Further, the planning module enables selecting an advertising impression goal for advertisement, assigning an advertising type and defining a weight for the advertisements. The control module receives the schedule, the advertising type and the defined weights and generates one or more metadata files that contain target criteria, advertising type and weights for the advertisements. The one or more metadata files, with the advertisements, are delivered to the receiver module that is configured to define a display frequency for the advertisements based upon one or more of the metadata files.Type: GrantFiled: June 16, 2004Date of Patent: January 11, 2011Assignee: Microsoft CorporationInventors: Stuart Ozer, Michael Patrick Hart, Wei Wei Ada Cho, Carolyn Khanh Chau
-
Patent number: 7757250Abstract: The present invention is related to a system and method of considering time segments or intervals in a collaborative filtering model. The present invention extends collaborative filtering approaches by integrating considerations of temporality into the training and/or vote input associated with the usage of collaborative filtering models. The present invention also applies filtering to the output with temporal models, so as to view a most appropriate subset of recommended content, centering on content that may be available at a target time. The present invention applies time to a collaborative filtering model by allowing weight to be associated with selections within a current time segment, selections historically watched within the current time segment by the user and selections historically watched within the current time segment by a large group of users.Type: GrantFiled: April 4, 2001Date of Patent: July 13, 2010Assignee: Microsoft CorporationInventors: Eric J. Horvitz, Carl M. Kadie, Stuart Ozer
-
Patent number: 7647365Abstract: A system and method of caching data employing probabilistic predictive techniques that provides local storage of a subset of available viewing selections by assigning a value to a selection and retaining selections in the cache depending on the value and size of the selection. The value assigned to an item can represent the time-dependent likelihood that a user will review an item at some time in the future. An initial value of an item can be based on the user's viewing habits, the user's viewing habit over particular time segment and/or viewing habits of a group of user's during a particular time segment. A value assigned to a selection dynamically changes according to a set of cache retention policies, where the value can be time-dependent functions that decay based on the class of the item, as determined by inference about the class or via a label associated with the item.Type: GrantFiled: September 18, 2008Date of Patent: January 12, 2010Assignee: Microsoft CorporationInventors: Eric J. Horvitz, Carl M. Kadie, Stuart Ozer, Curtis G. Wong
-
Patent number: 7644427Abstract: The present invention is related to a system and method of considering time segments or intervals in a collaborative filtering model. The present invention extends collaborative filtering approaches by integrating considerations of temporality into the training and/or vote input associated with the usage of collaborative filtering models. The present invention also applies filtering to the output with temporal models, so as to view a most appropriate subset of recommended content, centering on content that may be available at a target time. The present invention applies time to a collaborative filtering model by allowing weight to be associated with selections within a current time segment, selections historically watched within the current time segment by the user and selections historically watched within the current time segment by a large group of users.Type: GrantFiled: January 31, 2005Date of Patent: January 5, 2010Assignee: Microsoft CorporationInventors: Eric J. Horvitz, Carl M. Kadie, Stuart Ozer
-
Patent number: 7536316Abstract: A system, including a planning module, a control module and a receiver module, configured to schedule display of one or more advertising impressions of available advertising inventory. The planning module enables scheduling a requested quantity of advertising impressions in accordance with target criteria. Further, the planning module enables selecting an advertising impression goal for advertisement, assigning an advertising type and defining a weight for the advertisements. The control module receives the schedule, the advertising type and the defined weights and generates one or more metadata files that contain target criteria, advertising type and weights for the advertisements. The one or more metadata files, with the advertisements, are delivered to the receiver module that is configured to define a display frequency for the advertisements based upon one or more of the metadata files.Type: GrantFiled: June 14, 2004Date of Patent: May 19, 2009Assignee: Microsoft CorporationInventors: Stuart Ozer, Michael Patrick Hart, Wei Wei Ada Cho, Carolyn Khanh Chau
-
Publication number: 20090030925Abstract: Data clustering is described for determining related components of the data. In one example, information on biomolecules may be clustered into groups in which the biomolecular data in a clustered group may indicate data that shares a relationship. For example, monomers of a protein molecule or a nucleic acid molecule may be mapped to an evolutionary or phylogenetic tree. Candidate groupings of the information may be obtained based on evolutionary relationships among sequences corresponding to the molecules.Type: ApplicationFiled: July 23, 2007Publication date: January 29, 2009Applicant: Microsoft CorporationInventor: Stuart Ozer