Patents by Inventor Karen Stark

Karen Stark 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: 10402748
    Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.
    Type: Grant
    Filed: June 5, 2015
    Date of Patent: September 3, 2019
    Assignee: HEMANT V. VIRKAR
    Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
  • Publication number: 20160159132
    Abstract: The present invention relates to a completed book which comprises a front cover (1) with a folding crease (3), back cover (4) and spine (7) which form a covering member for copyrighted media products (9) wherein rivet screws (5) with closed sheaths (6) are inserted through punched holes (2) (8) to form a completed book. According to the invention, this process allows for a completed book to be made at will without special skills, training or the use of ancillary tools. This process does not use glue, a helical coil, staples or stitching.
    Type: Application
    Filed: October 29, 2015
    Publication date: June 9, 2016
    Inventor: Karen Starks
  • Publication number: 20150286955
    Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.
    Type: Application
    Filed: June 5, 2015
    Publication date: October 8, 2015
    Inventors: Hemant VIRKAR, Karen STARK, Jacob BORGMAN
  • Patent number: 9082083
    Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.
    Type: Grant
    Filed: January 22, 2013
    Date of Patent: July 14, 2015
    Assignee: DIGITAL INFUZION, INC.
    Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
  • Patent number: 8812274
    Abstract: Methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces, and to machines and systems relating thereto. More specifically, exemplary aspects of the invention relate to methods and systems for generating supervised hypersurfaces based on user domain expertise, machine learning techniques, or other supervised learning techniques. These supervised hypersurfaces may optionally be combined with unsupervised hypersurfaces derived from unsupervised learning techniques. Lower-dimensional subspaces may be determined by the methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces. Data may then be projected onto the lower-dimensional subspaces for use, e.g., in further data discovery, visualization for display, or database access.
    Type: Grant
    Filed: April 26, 2010
    Date of Patent: August 19, 2014
    Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
  • Publication number: 20130238533
    Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.
    Type: Application
    Filed: January 22, 2013
    Publication date: September 12, 2013
    Applicant: Digital Infuzion, Inc.
    Inventors: Hemant VIRKAR, Karen Stark, Jacob Borgman
  • Patent number: 8386401
    Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.
    Type: Grant
    Filed: September 10, 2009
    Date of Patent: February 26, 2013
    Assignee: Digital Infuzion, Inc.
    Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
  • Publication number: 20100274539
    Abstract: Methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces, and to machines and systems relating thereto. More specifically, exemplary aspects of the invention relate to methods and systems for generating supervised hypersurfaces based on user domain expertise, machine learning techniques, or other supervised learning techniques. These supervised hypersurfaces may optionally be combined with unsupervised hypersurfaces derived from unsupervised learning techniques. Lower-dimensional subspaces may be determined by the methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces. Data may then be projected onto the lower-dimensional subspaces for use, e.g., in further data discovery, visualization for display, or database access.
    Type: Application
    Filed: April 26, 2010
    Publication date: October 28, 2010
    Inventors: Hemant VIRKAR, Karen STARK, Jacob BORGMAN
  • Publication number: 20100063948
    Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.
    Type: Application
    Filed: September 10, 2009
    Publication date: March 11, 2010
    Applicant: DIGITAL INFUZION, INC.
    Inventors: Hemant VIRKAR, Karen Stark, Jacob Borgman
  • Publication number: 20040249145
    Abstract: The present invention provides multiple polynucleotide sequences from the same novel gene, the exons comprising the polynucleotide sequences, and the proteins encoded by the polynucleotide sequences. Three splicing variant polynucleotides were isolated from prostate tissue. The polypeptides, including the splicing variants, have a region of hydrophobicity indicative of a transmembrane domain and all three extracellular and cytoplasmic domains.
    Type: Application
    Filed: November 7, 2003
    Publication date: December 9, 2004
    Inventors: Karen A. Stark, Alix Weaver, Heidi M. Hoffmann, Raul Krauss, Kulvinder Singh Saini, Dario B. Valenzuela