TENGraD: Time-Efficient Natural Gradient Descent with Exact Fisher-Block Inversion

This work proposes a time-efficient Natural Gradient Descent method, called TENGraD, with linear convergence guarantees. Computing the inverse of the neural network's Fisher information matrix is expensive in NGD because the Fisher matrix is large. Approximate NGD methods such as KFAC attempt t...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2022-03
Main Authors: Soori, Saeed, Can, Bugra, Mu, Baourun, Gürbüzbalaban, Mert, Maryam Mehri Dehnavi
Format: Article
Language:English
Subjects:
Approximation
Convergence
Curvature
Image classification
Mathematical analysis
Matrix methods
Neural networks
Run time (computers)
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This work proposes a time-efficient Natural Gradient Descent method, called TENGraD, with linear convergence guarantees. Computing the inverse of the neural network's Fisher information matrix is expensive in NGD because the Fisher matrix is large. Approximate NGD methods such as KFAC attempt to improve NGD's running time and practical application by reducing the Fisher matrix inversion cost with approximation. However, the approximations do not reduce the overall time significantly and lead to less accurate parameter updates and loss of curvature information. TENGraD improves the time efficiency of NGD by computing Fisher block inverses with a computationally efficient covariance factorization and reuse method. It computes the inverse of each block exactly using the Woodbury matrix identity to preserve curvature information while admitting (linear) fast convergence rates. Our experiments on image classification tasks for state-of-the-art deep neural architecture on CIFAR-10, CIFAR-100, and Fashion-MNIST show that TENGraD significantly outperforms state-of-the-art NGD methods and often stochastic gradient descent in wall-clock time.
ISSN:2331-8422