Patents by Inventor Oleksii Tosov

Oleksii Tosov 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: 12088502
    Abstract: A method is provided for rate limiting in a network. The method comprises receiving a traffic flow from the network. In a supervised learning phase, and determining if the traffic flow matches a pretrained network flow model. If so, the method comprises designating the traffic flow as a classified traffic flow according to the pretrained network flow model. The method further comprises advancing to a grouping phase, conditioned upon the traffic flow not matching pretrained network flow models. In the unsupervised learning phase, the method comprises designating the traffic flow as a classified traffic flow. In the grouping phase that follows the supervised learning phase and the unsupervised learning phase, the method comprises using side information about the traffic flows to assign related traffic flows into traffic flow groups, identifying a particular traffic flow group from the traffic flow groups as being an offending traffic flow group.
    Type: Grant
    Filed: December 9, 2021
    Date of Patent: September 10, 2024
    Assignee: International Business Machines Corporation
    Inventors: Michal Paluch, Oleksii Tosov, Szymon Kowalczyk, Marcel Butucea Panait
  • Publication number: 20230188470
    Abstract: A method is provided for rate limiting in a network. The method comprises receiving a traffic flow from the network. In a supervised learning phase, and determining if the traffic flow matches a pretrained network flow model. If so, the method comprises designating the traffic flow as a classified traffic flow according to the pretrained network flow model. The method further comprises advancing to a grouping phase, conditioned upon the traffic flow not matching pretrained network flow models. In the unsupervised learning phase, the method comprises designating the traffic flow as a classified traffic flow. In the grouping phase that follows the supervised learning phase and the unsupervised learning phase, the method comprises using side information about the traffic flows to assign related traffic flows into traffic flow groups, identifying a particular traffic flow group from the traffic flow groups as being an offending traffic flow group.
    Type: Application
    Filed: December 9, 2021
    Publication date: June 15, 2023
    Inventors: Michal Paluch, Oleksii Tosov, Szymon Kowalczyk, Marcel Butucea Panait