Patents by Inventor Aakash Srinivasan

Aakash Srinivasan 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).

  • Utilizing a deep generative model with task embedding for personalized targeting of digital content through multiple channels across client devices
    Patent number: 11109083
    Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for training and utilizing a generative machine learning model to select one or more treatments for a client device from a set of treatments based on digital characteristics corresponding to the client device. In particular, the disclosed systems can train and apply a variational autoencoder with a task embedding layer that generates estimated effects for treatment combinations. For example, the disclosed systems receive, as input, digital characteristics corresponding to the client device and various treatment combinations. The disclosed systems apply the trained generative machine learning model with the task embedding layer to the digital characteristics to generate effect estimations for the various treatment combinations. Based on the effect estimations for the treatment combinations, the disclosed systems select one or more treatments to provide to the client device.
    Type: Grant
    Filed: January 25, 2019
    Date of Patent: August 31, 2021
    Assignee: ADOBE INC.
    Inventors: Shiv Kumar Saini, Sunny Dhamnani, Prithviraj Abasaheb Chavan, A S Akil Arif Ibrahim, Aakash Srinivasan
  • UTILIZING A DEEP GENERATIVE MODEL WITH TASK EMBEDDING FOR PERSONALIZED TARGETING OF DIGITAL CONTENT THROUGH MULTIPLE CHANNELS ACROSS CLIENT DEVICES
    Publication number: 20200245009
    Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for training and utilizing a generative machine learning model to select one or more treatments for a client device from a set of treatments based on digital characteristics corresponding to the client device. In particular, the disclosed systems can train and apply a variational autoencoder with a task embedding layer that generates estimated effects for treatment combinations. For example, the disclosed systems receive, as input, digital characteristics corresponding to the client device and various treatment combinations. The disclosed systems apply the trained generative machine learning model with the task embedding layer to the digital characteristics to generate effect estimations for the various treatment combinations. Based on the effect estimations for the treatment combinations, the disclosed systems select one or more treatments to provide to the client device.
    Type: Application
    Filed: January 25, 2019
    Publication date: July 30, 2020
    Inventors: Shiv Kumar Saini, Sunny Dhamnani, Prithviraj Abasaheb Chavan, AS Akil Arif Ibrahim, Aakash Srinivasan