Train Like a (Var)Pro: Efficient Training of Neural Networks with Variable Projection

TL;DR

This paper introduces GNvpro, an extension of variable projection for neural network training, which improves efficiency and generalization over traditional stochastic gradient descent methods in various applications.

Contribution

The paper develops the Gauss-Newton VarPro method for training DNNs with non-quadratic loss functions, broadening the applicability of variable projection techniques.

Findings

GNvpro outperforms SGD in efficiency across multiple tasks.

GNvpro achieves better generalization to unseen data.

Applicable to DNNs with affine last layers in classification tasks.

Abstract

Deep neural networks (DNNs) have achieved state-of-the-art performance across a variety of traditional machine learning tasks, e.g., speech recognition, image classification, and segmentation. The ability of DNNs to efficiently approximate high-dimensional functions has also motivated their use in scientific applications, e.g., to solve partial differential equations (PDE) and to generate surrogate models. In this paper, we consider the supervised training of DNNs, which arises in many of the above applications. We focus on the central problem of optimizing the weights of the given DNN such that it accurately approximates the relation between observed input and target data. Devising effective solvers for this optimization problem is notoriously challenging due to the large number of weights, non-convexity, data-sparsity, and non-trivial choice of hyperparameters. To solve the optimization problem more efficiently, we propose the use of variable projection (VarPro), a method originally designed for separable nonlinear least-squares problems. Our main contribution is the Gauss-Newton VarPro method (GNvpro) that extends the reach of the VarPro idea to non-quadratic objective functions, most notably, cross-entropy loss functions arising in classification. These extensions make GNvpro applicable to all training problems that involve a DNN whose last layer is an affine mapping, which is common in many state-of-the-art architectures. In our four numerical experiments from surrogate modeling, segmentation, and classification GNvpro solves the optimization problem more efficiently than commonly-used stochastic gradient descent (SGD) schemes. Also, GNvpro finds solutions that generalize well, and in all but one example better than well-tuned SGD methods, to unseen data points.