The Effect of Data Dimensionality on Neural Network Prunability
This addresses a problem for practitioners seeking to prune neural networks efficiently, but it appears incremental as it builds on prior work linking data structure to learning efficiency.
The paper investigates how the low-dimensional structure of high-dimensional input data, such as images and text, affects the prunability of neural networks, which is the maximum fraction of weights that can be pruned without losing test accuracy, but does not report specific numerical results.
Practitioners prune neural networks for efficiency gains and generalization improvements, but few scrutinize the factors determining the prunability of a neural network the maximum fraction of weights that pruning can remove without compromising the model's test accuracy. In this work, we study the properties of input data that may contribute to the prunability of a neural network. For high dimensional input data such as images, text, and audio, the manifold hypothesis suggests that these high dimensional inputs approximately lie on or near a significantly lower dimensional manifold. Prior work demonstrates that the underlying low dimensional structure of the input data may affect the sample efficiency of learning. In this paper, we investigate whether the low dimensional structure of the input data affects the prunability of a neural network.