Patents by Inventor Ioana Laura MARGINAS

Ioana Laura MARGINAS 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).

  • Machine learning system for adjusting operational characteristics of a computing system based upon HID activity
    Patent number: 11209805
    Abstract: Technologies are described for utilizing machine learning (“ML”) to adjust operational characteristics of a computing system based upon detected HID activity. Labeled training data is collected with user consent that includes data describing HID activity and data that identifies user activity taking place on a computing device when the data HID activity took place. A ML model is trained using the labeled training data that can receive data describing current HID activity and identify user activity currently taking place on another computing device based upon the current HID activity. The ML model can then select features of the other computing device that are beneficial to the identified user activity. The ML model can then cause one or more operational characteristics of the other computing device to be adjusted based upon the identified user activity, thereby saving valuable computing resources. A UI can also be presented that describes the identified features.
    Type: Grant
    Filed: October 31, 2017
    Date of Patent: December 28, 2021
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC.
    Inventors: Xiaoyu Chai, Choo Yei Chong, Ioana Laura Marginas, Eleanor Ann Robinson, Dale R. Johnson, Xinyi Zhang, Xiao Cai
  • MACHINE LEARNING SYSTEM FOR ADJUSTING OPERATIONAL CHARACTERISTICS OF A COMPUTING SYSTEM BASED UPON HID ACTIVITY
    Publication number: 20190129401
    Abstract: Technologies are described for utilizing machine learning (“ML”) to adjust operational characteristics of a computing system based upon detected HID activity. Labeled training data is collected with user consent that includes data describing HID activity and data that identifies user activity taking place on a computing device when the data HID activity took place. A ML model is trained using the labeled training data that can receive data describing current HID activity and identify user activity currently taking place on another computing device based upon the current HID activity. The ML model can then select features of the other computing device that are beneficial to the identified user activity. The ML model can then cause one or more operational characteristics of the other computing device to be adjusted based upon the identified user activity, thereby saving valuable computing resources. A UI can also be presented that describes the identified features.
    Type: Application
    Filed: October 31, 2017
    Publication date: May 2, 2019
    Inventors: Xiaoyu CHAI, Choo Yei CHONG, Ioana Laura MARGINAS, Eleanor Ann ROBINSON, Dale R. JOHNSON, Xinyi ZHANG, Xiao CAI