Patents by Inventor Muhammad AMJAD

Muhammad AMJAD 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: 11775608
    Abstract: A system and method model a time series from missing data by imputing missing values, denoising measured but noisy values, and forecasting future values of a single time series. A time series of potentially noisy, partially-measured values of a physical process is represented as a non-overlapping matrix. For several classes of common model functions, it can be proved that the resulting matrix has a low rank or approximately low rank, allowing a matrix estimation technique, for example singular value thresholding, to be efficiently applied. Applying such a technique produces a mean matrix that estimates latent values, of the physical process at times or intervals corresponding to measurements, with less error than previously known methods. These latent values have been denoised (if noisy) and imputed (if missing). Linear regression of the estimated latent values permits forecasting with an error that decreases as more measurements are made.
    Type: Grant
    Filed: July 7, 2022
    Date of Patent: October 3, 2023
    Assignee: Massachusetts Institute of Technology
    Inventors: Devavrat D. Shah, Anish Agarwal, Muhammad Amjad, Dennis Shen
  • Publication number: 20220366009
    Abstract: A system and method model a time series from missing data by imputing missing values, denoising measured but noisy values, and forecasting future values of a single time series. A time series of potentially noisy, partially-measured values of a physical process is represented as a non-overlapping matrix. For several classes of common model functions, it can be proved that the resulting matrix has a low rank or approximately low rank, allowing a matrix estimation technique, for example singular value thresholding, to be efficiently applied. Applying such a technique produces a mean matrix that estimates latent values, of the physical process at times or intervals corresponding to measurements, with less error than previously known methods. These latent values have been denoised (if noisy) and imputed (if missing). Linear regression of the estimated latent values permits forecasting with an error that decreases as more measurements are made.
    Type: Application
    Filed: July 7, 2022
    Publication date: November 17, 2022
    Applicant: Massachusetts Institute of Technology
    Inventors: Devavrat D. SHAH, Anish AGARWAL, Muhammad AMJAD, Dennis SHEN
  • Patent number: 11423118
    Abstract: A system and method model a time series from missing data by imputing missing values, denoising measured but noisy values, and forecasting future values of a single time series. A time series of potentially noisy, partially-measured values of a physical process is represented as a non-overlapping matrix. For several classes of common model functions, it can be proved that the resulting matrix has a low rank or approximately low rank, allowing a matrix estimation technique, for example singular value thresholding, to be efficiently applied. Applying such a technique produces a mean matrix that estimates latent values, of the physical process at times or intervals corresponding to measurements, with less error than previously known methods. These latent values have been denoised (if noisy) and imputed (if missing). Linear regression of the estimated latent values permits forecasting with an error that decreases as more measurements are made.
    Type: Grant
    Filed: January 7, 2019
    Date of Patent: August 23, 2022
    Assignee: Massachusetts Institute of Technology
    Inventors: Devavrat D. Shah, Anish Agarwal, Muhammad Amjad, Dennis Shen
  • Publication number: 20200218776
    Abstract: A system and method model a time series from missing data by imputing missing values, denoising measured but noisy values, and forecasting future values of a single time series. A time series of potentially noisy, partially-measured values of a physical process is represented as a non-overlapping matrix. For several classes of common model functions, it can be proved that the resulting matrix has a low rank or approximately low rank, allowing a matrix estimation technique, for example singular value thresholding, to be efficiently applied. Applying such a technique produces a mean matrix that estimates latent values, of the physical process at times or intervals corresponding to measurements, with less error than previously known methods. These latent values have been denoised (if noisy) and imputed (if missing). Linear regression of the estimated latent values permits forecasting with an error that decreases as more measurements are made.
    Type: Application
    Filed: January 7, 2019
    Publication date: July 9, 2020
    Inventors: Devavrat D. SHAH, Anish AGARWAL, Muhammad AMJAD, Dennis SHEN