On the Variance of the Fisher Information for Deep Learning

Alexander Soen, Ke Sun

Keywords: Fisher information, natural gradient, Cramer-Rao Lower Bound, deep learning

TL;DR: We explore the variance of the Fisher information matrix in the context of deep learning.

Abstract: In the realm of deep learning, the Fisher information matrix (FIM) gives novel insights and useful tools to characterize the loss landscape, perform second-order optimization, and build geometric learning theories. The exact FIM is either unavailable in closed form or too expensive to compute. In practice, it is almost always estimated based on empirical samples. We investigate two such estimators based on two equivalent representations of the FIM --- both unbiased and consistent. Their estimation quality is naturally gauged by their variance given in closed form. We analyze how the parametric structure of a deep neural network can affect the variance. The meaning of this variance measure and its upper bounds are then discussed in the context of deep learning.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2107.04205/code)