Skip to main content

Colloquium: Dr. Qiang Ye

Dr. Qiang Ye
University of Kentucky
Batch Normalization and Preconditioning for Neural Network Training
Wednesday, April 6, 2022
2:30 - 3:20 p.m.
Wallace Bldg. Room 348

Zoom: https://eku.zoom.us/j/99930025945


ABSTRACT

Batch normalization (BN) is a popular method in deep neural network training that
has been shown to decrease training time and improve generalization performance.
Despite its success, BN is not theoretically well understood. It is not suitable for use
with very small mini-batch sizes or online learning. In this talk, we will review BN and
present a preconditioning method called Batch Normalization Preconditioning (BNP)
to accelerate neural network training. We will analyze the effects of mini-batch statistics
of a hidden variable on the Hessian matrix of a loss function and propose a parameter
transformation that is equivalent to normalizing the hidden variables to improve
the conditioning of the Hessian. Compared with BN, one benefit of BNP is that it is not
constrained on the mini-batch size and works in the online learning setting. We will
present several experiments demonstrating competitiveness of BNP. Furthermore,
we will discuss a connection to BN which provides theoretical insights on how BN
improves training and how BN is applied to special architectures such as convolutional
neural networks. The talk is based on a joint work with Susanna Lange
and Kyle Helfrich.

Published on March 31, 2022

Open /*deleted href=#openmobile*/