Learning both Weights and Connections for Efficient Neural Networks

TL;DR

This paper introduces a method to significantly reduce neural network size and computation by learning and pruning important connections, enabling efficient deployment without accuracy loss.

Contribution

It presents a three-step pruning and retraining approach that automatically learns which connections are essential during training.

Findings

Reduced AlexNet parameters by 9x on ImageNet

Decreased VGG-16 parameters by 13x without accuracy loss

Achieved order-of-magnitude efficiency improvements

Abstract

Neural networks are both computationally intensive and memory intensive, making them difficult to deploy on embedded systems. Also, conventional networks fix the architecture before training starts; as a result, training cannot improve the architecture. To address these limitations, we describe a method to reduce the storage and computation required by neural networks by an order of magnitude without affecting their accuracy by learning only the important connections. Our method prunes redundant connections using a three-step method. First, we train the network to learn which connections are important. Next, we prune the unimportant connections. Finally, we retrain the network to fine tune the weights of the remaining connections. On the ImageNet dataset, our method reduced the number of parameters of AlexNet by a factor of 9x, from 61 million to 6.7 million, without incurring accuracy loss. Similar experiments with VGG-16 found that the number of parameters can be reduced by 13x, from 138 million to 10.3 million, again with no loss of accuracy.