On the Learning Dynamics of Two-layer Nonlinear Convolutional Neural Networks

TL;DR

This paper investigates the learning dynamics of a two-layer nonlinear convolutional neural network, revealing how filters learn key data patterns and achieve perfect accuracy under certain conditions, with implications for understanding CNN success.

Contribution

It provides a theoretical and empirical analysis of CNN learning dynamics on realistic data distributions, highlighting the emergence of key pattern filters and their dominance during training.

Findings

Filters learn key data patterns during training.

Filter norms dominate as training progresses.

CNN can achieve 100% accuracy on specific data distributions.

Abstract

Convolutional neural networks (CNNs) have achieved remarkable performance in various fields, particularly in the domain of computer vision. However, why this architecture works well remains to be a mystery. In this work we move a small step toward understanding the success of CNNs by investigating the learning dynamics of a two-layer nonlinear convolutional neural network over some specific data distributions. Rather than the typical Gaussian assumption for input data distribution, we consider a more realistic setting that each data point (e.g. image) contains a specific pattern determining its class label. Within this setting, we both theoretically and empirically show that some convolutional filters will learn the key patterns in data and the norm of these filters will dominate during the training process with stochastic gradient descent. And with any high probability, when the number of iterations is sufficiently large, the CNN model could obtain 100% accuracy over the considered data distributions. Our experiments demonstrate that for practical image classification tasks our findings still hold to some extent.