Patents by Inventor Deepika Bablani

Deepika Bablani 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).

  • Learned step size quantization
    Patent number: 11823054
    Abstract: Learned step size quantization in artificial neural network is provided. In various embodiments, a system comprises an artificial neural network and a computing node. The artificial neural network comprises: a quantizer having a configurable step size, the quantizer adapted to receive a plurality of input values and quantize the plurality of input values according to the configurable step size to produce a plurality of quantized input values, at least one matrix multiplier configured to receive the plurality of quantized input values from the quantizer and to apply a plurality of weights to the quantized input values to determine a plurality of output values having a first precision, and a multiplier configured to scale the output values to a second precision.
    Type: Grant
    Filed: February 20, 2020
    Date of Patent: November 21, 2023
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Steve Esser, Jeffrey L. McKinstry, Deepika Bablani, Rathinakumar Appuswamy, Dharmendra S. Modha
  • LEARNED STEP SIZE QUANTIZATION
    Publication number: 20210264279
    Abstract: Learned step size quantization in artificial neural network is provided. In various embodiments, a system comprises an artificial neural network and a computing node. The artificial neural network comprises: a quantizer having a configurable step size, the quantizer adapted to receive a plurality of input values and quantize the plurality of input values according to the configurable step size to produce a plurality of quantized input values, at least one matrix multiplier configured to receive the plurality of quantized input values from the quantizer and to apply a plurality of weights to the quantized input values to determine a plurality of output values having a first precision, and a multiplier configured to scale the output values to a second precision.
    Type: Application
    Filed: February 20, 2020
    Publication date: August 26, 2021
    Inventors: Steve Esser, Jeffrey L. McKinstry, Deepika Bablani, Rathinakumar Appuswamy, Dharmendra S. Modha