An empirical study of the relation between network architecture and complexity
This is an incremental study that addresses the problem of understanding network architecture effects for researchers in machine learning.
The authors propose an empirical study to investigate how network capacity affects generalization performance as data complexity increases, specifically by measuring generalization error in an image classification task with a steadily increasing number of classes and comparing various modern architectures at different scales.
In this preregistration submission, we propose an empirical study of how networks handle changes in complexity of the data. We investigate the effect of network capacity on generalization performance in the face of increasing data complexity. For this, we measure the generalization error for an image classification task where the number of classes steadily increases. We compare a number of modern architectures at different scales in this setting. The methodology, setup, and hypotheses described in this proposal were evaluated by peer review before experiments were conducted.