QHM (Quasi-Hyperbolic Momentum)

Published on: August 13, 2021

QHM (Quasi-Hyperbolic Momentum)

Table of Content

Quasi-Hyperbolic Momentum Algorithm (QHM) is a simple alteration of SGD with momentum, averaging a plain SGD step with a momentum step, thereby decoupling the momentum term from the current gradient when updating the weights.

The authors recommend  and  as a good starting point. For more information about QHM, check out the resources below.



More stories

  • Kernel PCA

  • Principal Component Analysis (PCA)

  • Linear Discriminant Analysis (LDA)