Patents by Inventor Janardhan Kulkarni

Janardhan Kulkarni 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).

  • Workload-Aware Hardware Architecture Recommendations
    Publication number: 20240126611
    Abstract: The description relates to accelerator architectures for deep learning models. One example can obtain a deep learning training script associated with a deep learning model and extract an operator graph from the training script. The example can split the operator graph into first and second portions of a heterogeneous pipeline and tune a first accelerator core for the first portion of the heterogeneous pipeline and a second accelerator core for the second portion of the heterogeneous pipeline. The example can also generate a hardware architecture that includes the first accelerator core and the second accelerator core arranged to collectively accomplish the deep learning model.
    Type: Application
    Filed: October 13, 2022
    Publication date: April 18, 2024
    Applicant: Microsoft Technology Licensing, LLC
    Inventors: Amar PHANISHAYEE, Divya MAHAJAN, Janardhan KULKARNI, Miguel CASTRO, Muhammad ADNAN
  • Noise generation for differential privacy
    Patent number: 11343012
    Abstract: A system and method for applying noise to data is described. The system accesses a metric value of a metric of each user from a group of users of an application. The metric indicates a measure of an operation of the application by a corresponding user. The system generates noise values and defines a distribution of the noise values to the group of users. The system modifies the metric value of the metric of each user with a corresponding noise value from the noise values based on the distribution.
    Type: Grant
    Filed: March 5, 2020
    Date of Patent: May 24, 2022
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Anjaneya Malpani, Jagadeesh Virupaksha Huliyar, Xinyun Sun, Sreeram Nivarthi, Muthukrishnan Paramasivam, Dheepak Ramaswamy, Sriradha Selvaraj, Ananthatejas Raghavan, Sergey Yekhanin, Janardhan Kulkarni, Aleksey Ashikhmin, Sivakanth Gopi, Bingran Luo
  • NOISE GENERATION FOR DIFFERENTIAL PRIVACY
    Publication number: 20210281483
    Abstract: A system and method for applying noise to data is described. The system accesses a metric value of a metric of each user from a group of users of an application. The metric indicates a measure of an operation of the application by a corresponding user. The system generates noise values and defines a distribution of the noise values to the group of users. The system modifies the metric value of the metric of each user with a corresponding noise value from the noise values based on the distribution.
    Type: Application
    Filed: March 5, 2020
    Publication date: September 9, 2021
    Inventors: Anjaneya Malpani, Jagadeesh Virupaksha Huliyar, Xinyun Sun, Sreeram Nivarthi, Muthukrishnan Paramasivam, Dheepak Ramaswamy, Sriradha Selvaraj, Ananthatejas Raghavan, Sergey Yekhanin, Janardhan Kulkarni, Aleksey Ashikhmin, Sivakanth Gopi, Bingran Luo