Patents by Inventor Paul Hankes Drielsma

Paul Hankes Drielsma 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).

  • User-controllable learning of policies
    Patent number: 8423483
    Abstract: Various embodiments are directed to a computer implemented method for updating a policy that is enforced by a computer program. In one embodiment, a computer communicates, to a user, data regarding one or more decisions made by the program over a period of time according to a policy. Each decision is made on the particular policy in force at the time the decision is made. Policy data for the policy is stored in a machine readable format. The user feedback data indicative of feedback by the user regarding the one or more decisions is stored. The computer identifies and ranks one or more potential variations to the policy based on a score of an objective function for each potential variation. The computer communicates, to the user, one or more suggested modifications based on the ranking of the one or more potential variations to the policy. The computer modifies the policy data based on one or more selections by the user in response to the transmission of the one or more suggested modifications.
    Type: Grant
    Filed: May 16, 2009
    Date of Patent: April 16, 2013
    Assignees: Carnegie Mellon University, Wombat Security Technology, Inc.
    Inventors: Norman Sadeh-Koniecpol, Paul Hankes Drielsma, Lorrie Faith Candor, Patrick Kelley
  • USER-CONTROLLABLE LEARNING OF POLICIES
    Publication number: 20100036779
    Abstract: Various embodiments are directed to a computer implemented method for updating a policy that is enforced by a computer program. In one embodiment, a computer communicates, to a user, data regarding one or more decisions made by the program over a period of time according to a policy. Each decision is made on the particular policy in force at the time the decision is made. Policy data for the policy is stored in a machine readable format. The user feedback data indicative of feedback by the user regarding the one or more decisions is stored. The computer identifies and ranks one or more potential variations to the policy based on a score of an objective function for each potential variation. The computer communicates, to the user, one or more suggested modifications based on the ranking of the one or more potential variations to the policy. The computer modifies the policy data based on one or more selections by the user in response to the transmission of the one or more suggested modifications.
    Type: Application
    Filed: May 16, 2009
    Publication date: February 11, 2010
    Inventors: Norman Sadeh-Koniecpol, Patrick Gage Kelley, Lorrie Faith Cranor, Paul Hankes Drielsma