BMPQ: Bit-Gradient Sensitivity-Driven Mixed-Precision Quantization of DNNs from Scratch

Document Details

Document Type
Pub Defense Publication
Publication Date
Mar 14, 2022
Source ID
10.23919/date54114.2022.9774740

Entities

People

  • Massoud Pedram
  • Peter A. Beerel
  • Qirui Sun
  • Shikai Wang
  • Souvik Kundu

Organizations

  • Defense Advanced Research Projects Agency
  • National Science Foundation
  • University of Southern California

Tags

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks