Data-Parallel Neural Network Training via Nonlinearly Preconditioned Trust-Region Method

TL;DR

This paper introduces a nonlinear preconditioned trust-region method for data-parallel neural network training that achieves comparable accuracy to SGD and Adam without hyperparameter tuning, enabling efficient parallelization.

Contribution

The paper presents a novel APTS variant that uses nonlinear preconditioning in a trust-region framework for parallel neural network training, reducing hyperparameter tuning requirements.

Findings

Achieves comparable accuracy to SGD and Adam on MNIST and CIFAR-10.

Enables parallel training without hyperparameter tuning.

Reduces computational cost associated with hyperparameter tuning.

Abstract

Parallel training methods are increasingly relevant in machine learning (ML) due to the continuing growth in model and dataset sizes. We propose a variant of the Additively Preconditioned Trust-Region Strategy (APTS) for training deep neural networks (DNNs). The proposed APTS method utilizes a data-parallel approach to construct a nonlinear preconditioner employed in the nonlinear optimization strategy. In contrast to the common employment of Stochastic Gradient Descent (SGD) and Adaptive Moment Estimation (Adam), which are both variants of gradient descent (GD) algorithms, the APTS method implicitly adjusts the step sizes in each iteration, thereby removing the need for costly hyperparameter tuning. We demonstrate the performance of the proposed APTS variant using the MNIST and CIFAR-10 datasets. The results obtained indicate that the APTS variant proposed here achieves comparable validation accuracy to SGD and Adam, all while allowing for parallel training and obviating the need for expensive hyperparameter tuning.