Patents by Inventor Susan Theresa Dumais
Susan Theresa Dumais 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).
-
Publication number: 20240046087Abstract: This document relates to training of machine learning models. One example method involves providing a machine learning model having a first classification layer, a second classification layer, and an encoder that feeds into the first classification layer and the second classification layer. The example method also involves obtaining first training examples having explicit labels and second training examples having inferred labels. The inferred labels are based at least on actions associated with the second training examples. The example method also involves training the machine learning model using the first training examples and the second training examples using a training objective that considers first training loss of the first classification layer for the explicit labels and second training loss of the second classification layer for the inferred labels. The method also involves outputting a trained machine learning model having the encoder and the first classification layer.Type: ApplicationFiled: October 4, 2023Publication date: February 8, 2024Applicant: Microsoft Technology Licensing, LLCInventors: Subhabrata Mukherjee, Guoqing Zheng, Ahmed Awadalla, Milad Shokouhi, Susan Theresa Dumais, Kai Shu
-
Patent number: 11816566Abstract: This document relates to training of machine learning models. One example method involves providing a machine learning model having a first classification layer, a second classification layer, and an encoder that feeds into the first classification layer and the second classification layer. The example method also involves obtaining first training examples having explicit labels and second training examples having inferred labels. The inferred labels are based at least on actions associated with the second training examples. The example method also involves training the machine learning model using the first training examples and the second training examples using a training objective that considers first training loss of the first classification layer for the explicit labels and second training loss of the second classification layer for the inferred labels. The method also involves outputting a trained machine learning model having the encoder and the first classification layer.Type: GrantFiled: May 18, 2020Date of Patent: November 14, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Subhabrata Mukherjee, Guoqing Zheng, Ahmed Awadalla, Milad Shokouhi, Susan Theresa Dumais, Kai Shu
-
Patent number: 11636394Abstract: The present concepts relate to a differentiable user-item co-clustering (“DUICC”) model for recommendation and co-clustering. Users' interaction with items (e.g., content) may be centered around information co-clusters—groups of items and users that exhibit common consumption behavior. The DUICC model may learn fine-grained co-cluster structures of items and users based on their interaction data. The DUICC model can then leverage the learned latent co-cluster structures to calculate preference stores of the items for a user. The top scoring items may be presented to the user as recommendations.Type: GrantFiled: June 25, 2020Date of Patent: April 25, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Longqi Yang, Tobias Benjamin Schnabel, Paul Nathan Bennett, Susan Theresa Dumais
-
Publication number: 20210406761Abstract: The present concepts relate to a differentiable user-item co-clustering (“DUICC”) model for recommendation and co-clustering. Users' interaction with items (e.g., content) may be centered around information co-clusters—groups of items and users that exhibit common consumption behavior. The DUICC model may learn fine-grained co-cluster structures of items and users based on their interaction data. The DUICC model can then leverage the learned latent co-cluster structures to calculate preference stores of the items for a user. The top scoring items may be presented to the user as recommendations.Type: ApplicationFiled: June 25, 2020Publication date: December 30, 2021Applicant: Microsoft Technology Licensing, LLCInventors: Longqi Yang, Tobias Benjamin Schnabel, Paul Nathan Bennett, Susan Theresa Dumais
-
Publication number: 20210357747Abstract: This document relates to training of machine learning models. One example method involves providing a machine learning model having a first classification layer, a second classification layer, and an encoder that feeds into the first classification layer and the second classification layer. The example method also involves obtaining first training examples having explicit labels and second training examples having inferred labels. The inferred labels are based at least on actions associated with the second training examples. The example method also involves training the machine learning model using the first training examples and the second training examples using a training objective that considers first training loss of the first classification layer for the explicit labels and second training loss of the second classification layer for the inferred labels. The method also involves outputting a trained machine learning model having the encoder and the first classification layer.Type: ApplicationFiled: May 18, 2020Publication date: November 18, 2021Applicant: Microsoft Technology Licensing, LLCInventors: Subhabrata Mukherjee, Guoqing Zheng, Ahmed Awadalla, Milad Shokouhi, Susan Theresa Dumais, Kai Shu
-
Patent number: 10037367Abstract: Systems, methods, and computer storage media are provided for analyzing a large amount of social media data from a large population of social media users and constructing correlational data models between one or more events that occur within each user's timeline. Social media posts directed to personal experiences of a large number of social media users are extracted. Event timelines are generated for each of the social media users, based on their personal experiences. The event timelines are analyzed with a particular event of interest to measure correlations between events occurring within the timelines and the particular event of interest. Using the measured correlations, a correlational data model is thereby constructed. The correlational data model may be used for application to decision-making calculations by one or more systems in an active or passive manner.Type: GrantFiled: December 15, 2014Date of Patent: July 31, 2018Assignee: MICROSOFT TECHNOLOGY LICENSING, LLCInventors: Emre Mehmet Kiciman, Paul Nathan Bennett, Jaime Brooks Teevan, Susan Theresa Dumais
-
Publication number: 20160171063Abstract: Systems, methods, and computer storage media are provided for analyzing a large amount of social media data from a large population of social media users and constructing correlational data models between one or more events that occur within each user's timeline. Social media posts directed to personal experiences of a large number of social media users are extracted. Event timelines are generated for each of the social media users, based on their personal experiences. The event timelines are analyzed with a particular event of interest to measure correlations between events occurring within the timelines and the particular event of interest. Using the measured correlations, a correlational data model is thereby constructed. The correlational data model may be used for application to decision-making calculations by one or more systems in an active or passive manner.Type: ApplicationFiled: December 15, 2014Publication date: June 16, 2016Inventors: EMRE MEHMET KICIMAN, PAUL NATHAN BENNETT, JAIME BROOKS TEEVAN, SUSAN THERESA DUMAIS
-
Publication number: 20120041953Abstract: A latent topic labels text mining system and method to mine and analyze the content of textual data. Embodiments of the system and method are particularly well suited for use on microblog data to help people identify posts they want to read and to find people that they want to follow. Embodiments of the system and method use a modified Labeled LDA technique (called an L+LDA technique) that analyzes content using a combination of labeled and latent topics. The resultant data is assigned labels one of four labels to generate a lower-dimensional representation of the data that the individual words in a microblog post. This learned topic representation is used to characterize, summarize, filter, find, suggest, and compare the content of microblog posts. Embodiments of the system and method also include visualization techniques such as a tag cloud visualization that is used to visualize microblogging data.Type: ApplicationFiled: August 16, 2010Publication date: February 16, 2012Applicant: Microsoft CorporationInventors: Susan Theresa Dumais, Daniel Ramage, Daniel John Liebling, Steven Mark Drucker