On the Ineffectiveness of Variance Reduced Optimization for Deep Learning

Aaron Defazio

Abstract: The application of stochastic variance reduction to optimization has shown remarkable recent theoretical and practical success. The applicability of these techniques to the hard non-convex optimization problems encountered during training of modern deep neural networks is an open problem. We show that naive application of the SVRG technique and related approaches fail, and explore why.

Keywords: machine learning, optimization, variance reduction

TL;DR: The SVRG method fails on modern deep learning problems

Data: [MNIST](https://paperswithcode.com/dataset/mnist)

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:1812.04529/code)